| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.7751937984496124, |
| "eval_steps": 500, |
| "global_step": 700, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "completion_length": 73.265625, |
| "epoch": 0.0011074197120708748, |
| "grad_norm": 0.5076314806938171, |
| "kl": 0.0, |
| "learning_rate": 9.99375e-07, |
| "loss": -0.018259915290400386, |
| "reward": 2.2648561000823975, |
| "reward_std": 0.32521533221006393, |
| "rewards/GDino": 0.84943026304245, |
| "rewards/GIT": 0.5776679813861847, |
| "rewards/HPSv2": 0.2639656066894531, |
| "rewards/ORM": 0.5737921893596649, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -22.0, |
| "step": 1 |
| }, |
| { |
| "completion_length": 56.0, |
| "epoch": 0.0022148394241417496, |
| "grad_norm": 0.5364330410957336, |
| "kl": 0.001522064208984375, |
| "learning_rate": 9.9875e-07, |
| "loss": 0.00348748016403988, |
| "reward": 1.7680926322937012, |
| "reward_std": 0.41801488399505615, |
| "rewards/GDino": 0.6529064476490021, |
| "rewards/GIT": 0.19494981318712234, |
| "rewards/HPSv2": 0.24983596801757812, |
| "rewards/ORM": 0.6704004406929016, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.0, |
| "step": 2 |
| }, |
| { |
| "completion_length": 55.4375, |
| "epoch": 0.0033222591362126247, |
| "grad_norm": 0.5614722967147827, |
| "kl": 0.001556396484375, |
| "learning_rate": 9.98125e-07, |
| "loss": 0.01565772108733654, |
| "reward": 1.6570448875427246, |
| "reward_std": 0.3965621292591095, |
| "rewards/GDino": 0.6382372081279755, |
| "rewards/GIT": 0.37795570492744446, |
| "rewards/HPSv2": 0.24709796905517578, |
| "rewards/ORM": 0.3937540017068386, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -20.9375, |
| "step": 3 |
| }, |
| { |
| "completion_length": 65.34375, |
| "epoch": 0.004429678848283499, |
| "grad_norm": 2.5736770629882812, |
| "kl": 0.0016021728515625, |
| "learning_rate": 9.975e-07, |
| "loss": -0.0012893765233457088, |
| "reward": 2.061529755592346, |
| "reward_std": 0.4106704443693161, |
| "rewards/GDino": 0.7796730995178223, |
| "rewards/GIT": 0.43717896938323975, |
| "rewards/HPSv2": 0.24744796752929688, |
| "rewards/ORM": 0.5972296595573425, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -22.0, |
| "step": 4 |
| }, |
| { |
| "completion_length": 63.578125, |
| "epoch": 0.005537098560354375, |
| "grad_norm": 0.48238250613212585, |
| "kl": 0.001575469970703125, |
| "learning_rate": 9.968749999999999e-07, |
| "loss": 0.020129199139773846, |
| "reward": 1.5302643775939941, |
| "reward_std": 0.44902199506759644, |
| "rewards/GDino": 0.6246840953826904, |
| "rewards/GIT": 0.23608428239822388, |
| "rewards/HPSv2": 0.2453451156616211, |
| "rewards/ORM": 0.42415088415145874, |
| "self_certainty_semantic": -25.625, |
| "self_certainty_token": -22.1875, |
| "step": 5 |
| }, |
| { |
| "completion_length": 60.65625, |
| "epoch": 0.006644518272425249, |
| "grad_norm": 0.8221905827522278, |
| "kl": 0.001674652099609375, |
| "learning_rate": 9.9625e-07, |
| "loss": 0.0192068200558424, |
| "reward": 2.1602972745895386, |
| "reward_std": 0.23134037852287292, |
| "rewards/GDino": 0.783700168132782, |
| "rewards/GIT": 0.452057421207428, |
| "rewards/HPSv2": 0.274627685546875, |
| "rewards/ORM": 0.6499120593070984, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.0, |
| "step": 6 |
| }, |
| { |
| "completion_length": 65.453125, |
| "epoch": 0.007751937984496124, |
| "grad_norm": 0.433403879404068, |
| "kl": 0.0016021728515625, |
| "learning_rate": 9.956249999999999e-07, |
| "loss": 0.028950304724276066, |
| "reward": 1.7097668647766113, |
| "reward_std": 0.5880981385707855, |
| "rewards/GDino": 0.5914062708616257, |
| "rewards/GIT": 0.15753822773694992, |
| "rewards/HPSv2": 0.25023555755615234, |
| "rewards/ORM": 0.7105867862701416, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -21.75, |
| "step": 7 |
| }, |
| { |
| "completion_length": 74.90625, |
| "epoch": 0.008859357696566999, |
| "grad_norm": 0.41245806217193604, |
| "kl": 0.00152587890625, |
| "learning_rate": 9.95e-07, |
| "loss": -0.016540683340281248, |
| "reward": 1.785366177558899, |
| "reward_std": 0.39637817442417145, |
| "rewards/GDino": 0.7011832594871521, |
| "rewards/GIT": 0.3848375529050827, |
| "rewards/HPSv2": 0.2445659637451172, |
| "rewards/ORM": 0.45477938652038574, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -20.875, |
| "step": 8 |
| }, |
| { |
| "completion_length": 61.828125, |
| "epoch": 0.009966777408637873, |
| "grad_norm": 0.3924250602722168, |
| "kl": 0.001617431640625, |
| "learning_rate": 9.94375e-07, |
| "loss": 0.03069412149488926, |
| "reward": 2.0813064575195312, |
| "reward_std": 0.5435488224029541, |
| "rewards/GDino": 0.736801415681839, |
| "rewards/GIT": 0.32275132089853287, |
| "rewards/HPSv2": 0.26233673095703125, |
| "rewards/ORM": 0.759416937828064, |
| "self_certainty_semantic": -25.5625, |
| "self_certainty_token": -21.1875, |
| "step": 9 |
| }, |
| { |
| "completion_length": 62.796875, |
| "epoch": 0.01107419712070875, |
| "grad_norm": 0.5886948704719543, |
| "kl": 0.00164031982421875, |
| "learning_rate": 9.9375e-07, |
| "loss": -0.009089878294616938, |
| "reward": 1.8167259693145752, |
| "reward_std": 0.4427160769701004, |
| "rewards/GDino": 0.6997816860675812, |
| "rewards/GIT": 0.4742187559604645, |
| "rewards/HPSv2": 0.2480792999267578, |
| "rewards/ORM": 0.3946462571620941, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -21.125, |
| "step": 10 |
| }, |
| { |
| "completion_length": 64.09375, |
| "epoch": 0.012181616832779624, |
| "grad_norm": 0.6388463377952576, |
| "kl": 0.0016326904296875, |
| "learning_rate": 9.93125e-07, |
| "loss": -0.011163983959704638, |
| "reward": 2.250586152076721, |
| "reward_std": 0.29546695202589035, |
| "rewards/GDino": 0.7932291626930237, |
| "rewards/GIT": 0.5437096580862999, |
| "rewards/HPSv2": 0.25614356994628906, |
| "rewards/ORM": 0.657503753900528, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.0625, |
| "step": 11 |
| }, |
| { |
| "completion_length": 73.265625, |
| "epoch": 0.013289036544850499, |
| "grad_norm": 0.37963175773620605, |
| "kl": 0.001583099365234375, |
| "learning_rate": 9.925e-07, |
| "loss": 0.009535952471196651, |
| "reward": 1.8723560571670532, |
| "reward_std": 0.48824670910835266, |
| "rewards/GDino": 0.671429455280304, |
| "rewards/GIT": 0.4155814051628113, |
| "rewards/HPSv2": 0.2387409210205078, |
| "rewards/ORM": 0.5466042459011078, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.625, |
| "step": 12 |
| }, |
| { |
| "completion_length": 55.015625, |
| "epoch": 0.014396456256921373, |
| "grad_norm": 0.5844080448150635, |
| "kl": 0.001674652099609375, |
| "learning_rate": 9.91875e-07, |
| "loss": 0.0034986711107194424, |
| "reward": 1.7595484256744385, |
| "reward_std": 0.3697086051106453, |
| "rewards/GDino": 0.7100214958190918, |
| "rewards/GIT": 0.26869260519742966, |
| "rewards/HPSv2": 0.24958419799804688, |
| "rewards/ORM": 0.53125, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.375, |
| "step": 13 |
| }, |
| { |
| "completion_length": 55.65625, |
| "epoch": 0.015503875968992248, |
| "grad_norm": 0.5192797780036926, |
| "kl": 0.001674652099609375, |
| "learning_rate": 9.912499999999998e-07, |
| "loss": 0.010001872200518847, |
| "reward": 2.201015591621399, |
| "reward_std": 0.4899330288171768, |
| "rewards/GDino": 0.8140625059604645, |
| "rewards/GIT": 0.4328514188528061, |
| "rewards/HPSv2": 0.2431640625, |
| "rewards/ORM": 0.7109375, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -22.0, |
| "step": 14 |
| }, |
| { |
| "completion_length": 64.0, |
| "epoch": 0.016611295681063124, |
| "grad_norm": 0.46844616532325745, |
| "kl": 0.00174713134765625, |
| "learning_rate": 9.90625e-07, |
| "loss": 0.0017675042618066072, |
| "reward": 2.433342456817627, |
| "reward_std": 0.33736473321914673, |
| "rewards/GDino": 0.9153576791286469, |
| "rewards/GIT": 0.5124611556529999, |
| "rewards/HPSv2": 0.2507901191711426, |
| "rewards/ORM": 0.7547334432601929, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.6875, |
| "step": 15 |
| }, |
| { |
| "completion_length": 53.203125, |
| "epoch": 0.017718715393133997, |
| "grad_norm": 0.49579355120658875, |
| "kl": 0.001758575439453125, |
| "learning_rate": 9.9e-07, |
| "loss": 0.003856237977743149, |
| "reward": 1.6368815302848816, |
| "reward_std": 0.42226114869117737, |
| "rewards/GDino": 0.6432631015777588, |
| "rewards/GIT": 0.2906690910458565, |
| "rewards/HPSv2": 0.25169944763183594, |
| "rewards/ORM": 0.45124977827072144, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -20.8125, |
| "step": 16 |
| }, |
| { |
| "completion_length": 76.28125, |
| "epoch": 0.018826135105204873, |
| "grad_norm": 0.5296036601066589, |
| "kl": 0.001590728759765625, |
| "learning_rate": 9.89375e-07, |
| "loss": -0.003345506265759468, |
| "reward": 1.7861530184745789, |
| "reward_std": 0.5057752877473831, |
| "rewards/GDino": 0.6293700635433197, |
| "rewards/GIT": 0.2197464406490326, |
| "rewards/HPSv2": 0.26516151428222656, |
| "rewards/ORM": 0.671875, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -20.6875, |
| "step": 17 |
| }, |
| { |
| "completion_length": 58.0625, |
| "epoch": 0.019933554817275746, |
| "grad_norm": 0.6577962636947632, |
| "kl": 0.00174713134765625, |
| "learning_rate": 9.8875e-07, |
| "loss": -0.019500677473843098, |
| "reward": 2.303292393684387, |
| "reward_std": 0.2609405145049095, |
| "rewards/GDino": 0.8339102566242218, |
| "rewards/GIT": 0.5853700041770935, |
| "rewards/HPSv2": 0.24338722229003906, |
| "rewards/ORM": 0.640625, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.25, |
| "step": 18 |
| }, |
| { |
| "completion_length": 50.859375, |
| "epoch": 0.021040974529346623, |
| "grad_norm": 0.3543226718902588, |
| "kl": 0.00182342529296875, |
| "learning_rate": 9.88125e-07, |
| "loss": -0.00019507110118865967, |
| "reward": 1.6344053149223328, |
| "reward_std": 0.47374215722084045, |
| "rewards/GDino": 0.705148845911026, |
| "rewards/GIT": 0.2559727430343628, |
| "rewards/HPSv2": 0.2541370391845703, |
| "rewards/ORM": 0.41914665699005127, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.3125, |
| "step": 19 |
| }, |
| { |
| "completion_length": 65.921875, |
| "epoch": 0.0221483942414175, |
| "grad_norm": 0.5358290672302246, |
| "kl": 0.001781463623046875, |
| "learning_rate": 9.875e-07, |
| "loss": 0.007933363318443298, |
| "reward": 1.9504321217536926, |
| "reward_std": 0.3728322237730026, |
| "rewards/GDino": 0.6606760025024414, |
| "rewards/GIT": 0.48046815395355225, |
| "rewards/HPSv2": 0.24678802490234375, |
| "rewards/ORM": 0.5625, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.0625, |
| "step": 20 |
| }, |
| { |
| "completion_length": 59.3125, |
| "epoch": 0.023255813953488372, |
| "grad_norm": 2.0912797451019287, |
| "kl": 0.001811981201171875, |
| "learning_rate": 9.86875e-07, |
| "loss": -0.004398644436150789, |
| "reward": 2.252086877822876, |
| "reward_std": 0.44888848066329956, |
| "rewards/GDino": 0.798213005065918, |
| "rewards/GIT": 0.4853799045085907, |
| "rewards/HPSv2": 0.25956153869628906, |
| "rewards/ORM": 0.7089323997497559, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.75, |
| "step": 21 |
| }, |
| { |
| "completion_length": 52.265625, |
| "epoch": 0.024363233665559248, |
| "grad_norm": 0.5790585875511169, |
| "kl": 0.00191497802734375, |
| "learning_rate": 9.862499999999999e-07, |
| "loss": 0.006876260507851839, |
| "reward": 1.9933909177780151, |
| "reward_std": 0.32367050647735596, |
| "rewards/GDino": 0.7134387493133545, |
| "rewards/GIT": 0.41087181866168976, |
| "rewards/HPSv2": 0.2721214294433594, |
| "rewards/ORM": 0.5969589203596115, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -22.375, |
| "step": 22 |
| }, |
| { |
| "completion_length": 59.375, |
| "epoch": 0.02547065337763012, |
| "grad_norm": 0.45692723989486694, |
| "kl": 0.001697540283203125, |
| "learning_rate": 9.85625e-07, |
| "loss": -0.00792664848268032, |
| "reward": 2.015365242958069, |
| "reward_std": 0.48256243765354156, |
| "rewards/GDino": 0.724082350730896, |
| "rewards/GIT": 0.42729710042476654, |
| "rewards/HPSv2": 0.2667560577392578, |
| "rewards/ORM": 0.5972296893596649, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -20.6875, |
| "step": 23 |
| }, |
| { |
| "completion_length": 55.203125, |
| "epoch": 0.026578073089700997, |
| "grad_norm": 0.46439889073371887, |
| "kl": 0.0016937255859375, |
| "learning_rate": 9.849999999999999e-07, |
| "loss": 0.0024933242239058018, |
| "reward": 2.460409939289093, |
| "reward_std": 0.4443647414445877, |
| "rewards/GDino": 0.8454739451408386, |
| "rewards/GIT": 0.6258784532546997, |
| "rewards/HPSv2": 0.2624950408935547, |
| "rewards/ORM": 0.7265625, |
| "self_certainty_semantic": -25.6875, |
| "self_certainty_token": -21.25, |
| "step": 24 |
| }, |
| { |
| "completion_length": 60.15625, |
| "epoch": 0.02768549280177187, |
| "grad_norm": 0.47176027297973633, |
| "kl": 0.001880645751953125, |
| "learning_rate": 9.84375e-07, |
| "loss": 0.005812188144773245, |
| "reward": 2.0174233317375183, |
| "reward_std": 0.40724658966064453, |
| "rewards/GDino": 0.7186038792133331, |
| "rewards/GIT": 0.4156235605478287, |
| "rewards/HPSv2": 0.26485633850097656, |
| "rewards/ORM": 0.6183395236730576, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -22.0, |
| "step": 25 |
| }, |
| { |
| "completion_length": 53.21875, |
| "epoch": 0.028792912513842746, |
| "grad_norm": 0.716375470161438, |
| "kl": 0.00209808349609375, |
| "learning_rate": 9.8375e-07, |
| "loss": 0.02397427149116993, |
| "reward": 2.186239182949066, |
| "reward_std": 0.46710920333862305, |
| "rewards/GDino": 0.7593750059604645, |
| "rewards/GIT": 0.5171153843402863, |
| "rewards/HPSv2": 0.2734565734863281, |
| "rewards/ORM": 0.6362921893596649, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -23.0, |
| "step": 26 |
| }, |
| { |
| "completion_length": 58.421875, |
| "epoch": 0.029900332225913623, |
| "grad_norm": 0.428893119096756, |
| "kl": 0.00171661376953125, |
| "learning_rate": 9.83125e-07, |
| "loss": -0.005866332910954952, |
| "reward": 1.9681838750839233, |
| "reward_std": 0.3645169883966446, |
| "rewards/GDino": 0.7666666209697723, |
| "rewards/GIT": 0.4486802965402603, |
| "rewards/HPSv2": 0.2419452667236328, |
| "rewards/ORM": 0.5108915567398071, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.4375, |
| "step": 27 |
| }, |
| { |
| "completion_length": 63.328125, |
| "epoch": 0.031007751937984496, |
| "grad_norm": 0.5334203243255615, |
| "kl": 0.002010345458984375, |
| "learning_rate": 9.825e-07, |
| "loss": 0.012586410157382488, |
| "reward": 1.4134111404418945, |
| "reward_std": 0.3155324012041092, |
| "rewards/GDino": 0.6005972325801849, |
| "rewards/GIT": 0.11092349141836166, |
| "rewards/HPSv2": 0.2596569061279297, |
| "rewards/ORM": 0.44223344326019287, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -20.5625, |
| "step": 28 |
| }, |
| { |
| "completion_length": 56.25, |
| "epoch": 0.03211517165005537, |
| "grad_norm": 0.40832045674324036, |
| "kl": 0.001819610595703125, |
| "learning_rate": 9.81875e-07, |
| "loss": 0.010300841182470322, |
| "reward": 2.465680956840515, |
| "reward_std": 0.298002652823925, |
| "rewards/GDino": 0.862500011920929, |
| "rewards/GIT": 0.6107669174671173, |
| "rewards/HPSv2": 0.28375244140625, |
| "rewards/ORM": 0.7086615860462189, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.0625, |
| "step": 29 |
| }, |
| { |
| "completion_length": 54.953125, |
| "epoch": 0.03322259136212625, |
| "grad_norm": 0.4050670266151428, |
| "kl": 0.002025604248046875, |
| "learning_rate": 9.8125e-07, |
| "loss": -0.001845305785536766, |
| "reward": 2.476737856864929, |
| "reward_std": 0.3756887763738632, |
| "rewards/GDino": 0.8967152833938599, |
| "rewards/GIT": 0.551719531416893, |
| "rewards/HPSv2": 0.24522781372070312, |
| "rewards/ORM": 0.7830752730369568, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.25, |
| "step": 30 |
| }, |
| { |
| "completion_length": 74.84375, |
| "epoch": 0.03433001107419712, |
| "grad_norm": 0.7089686393737793, |
| "kl": 0.001865386962890625, |
| "learning_rate": 9.806249999999998e-07, |
| "loss": 0.023707949556410313, |
| "reward": 1.831493854522705, |
| "reward_std": 0.37860143184661865, |
| "rewards/GDino": 0.6287499666213989, |
| "rewards/GIT": 0.3833145350217819, |
| "rewards/HPSv2": 0.2413043975830078, |
| "rewards/ORM": 0.578125, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -20.9375, |
| "step": 31 |
| }, |
| { |
| "completion_length": 71.0625, |
| "epoch": 0.035437430786267994, |
| "grad_norm": 0.45204266905784607, |
| "kl": 0.00200653076171875, |
| "learning_rate": 9.8e-07, |
| "loss": 0.014695112593472004, |
| "reward": 1.5279032588005066, |
| "reward_std": 0.5042913109064102, |
| "rewards/GDino": 0.6702238023281097, |
| "rewards/GIT": 0.24817809462547302, |
| "rewards/HPSv2": 0.2356252670288086, |
| "rewards/ORM": 0.37387609481811523, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.5625, |
| "step": 32 |
| }, |
| { |
| "completion_length": 59.703125, |
| "epoch": 0.036544850498338874, |
| "grad_norm": 0.4359590411186218, |
| "kl": 0.00201416015625, |
| "learning_rate": 9.79375e-07, |
| "loss": 0.00610552029684186, |
| "reward": 2.3108657598495483, |
| "reward_std": 0.4415571391582489, |
| "rewards/GDino": 0.8515625, |
| "rewards/GIT": 0.6067334115505219, |
| "rewards/HPSv2": 0.22726917266845703, |
| "rewards/ORM": 0.6253007054328918, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.5, |
| "step": 33 |
| }, |
| { |
| "completion_length": 58.046875, |
| "epoch": 0.03765227021040975, |
| "grad_norm": 0.5853399038314819, |
| "kl": 0.002033233642578125, |
| "learning_rate": 9.7875e-07, |
| "loss": 0.023541483096778393, |
| "reward": 2.012690246105194, |
| "reward_std": 0.4660336524248123, |
| "rewards/GDino": 0.6989582777023315, |
| "rewards/GIT": 0.40700431168079376, |
| "rewards/HPSv2": 0.24774932861328125, |
| "rewards/ORM": 0.6589783728122711, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.8125, |
| "step": 34 |
| }, |
| { |
| "completion_length": 56.90625, |
| "epoch": 0.03875968992248062, |
| "grad_norm": 0.3787715435028076, |
| "kl": 0.001888275146484375, |
| "learning_rate": 9.78125e-07, |
| "loss": 0.003942073322832584, |
| "reward": 2.452033281326294, |
| "reward_std": 0.3410096764564514, |
| "rewards/GDino": 0.8359375298023224, |
| "rewards/GIT": 0.567652553319931, |
| "rewards/HPSv2": 0.2418804168701172, |
| "rewards/ORM": 0.806562751531601, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.3125, |
| "step": 35 |
| }, |
| { |
| "completion_length": 66.0, |
| "epoch": 0.03986710963455149, |
| "grad_norm": 0.5305721163749695, |
| "kl": 0.005157470703125, |
| "learning_rate": 9.775e-07, |
| "loss": -0.003781900042667985, |
| "reward": 1.8618011474609375, |
| "reward_std": 0.4120703786611557, |
| "rewards/GDino": 0.6453125476837158, |
| "rewards/GIT": 0.4281370937824249, |
| "rewards/HPSv2": 0.24621009826660156, |
| "rewards/ORM": 0.5421415567398071, |
| "self_certainty_semantic": -25.5625, |
| "self_certainty_token": -20.9375, |
| "step": 36 |
| }, |
| { |
| "completion_length": 51.40625, |
| "epoch": 0.04097452934662237, |
| "grad_norm": 0.46515390276908875, |
| "kl": 0.002716064453125, |
| "learning_rate": 9.76875e-07, |
| "loss": 0.006902199704200029, |
| "reward": 1.9485998153686523, |
| "reward_std": 0.42147715389728546, |
| "rewards/GDino": 0.6951449513435364, |
| "rewards/GIT": 0.31057579815387726, |
| "rewards/HPSv2": 0.26158714294433594, |
| "rewards/ORM": 0.681291937828064, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.1875, |
| "step": 37 |
| }, |
| { |
| "completion_length": 71.03125, |
| "epoch": 0.042081949058693245, |
| "grad_norm": 0.951810896396637, |
| "kl": 0.00226593017578125, |
| "learning_rate": 9.7625e-07, |
| "loss": 0.03428783547133207, |
| "reward": 1.9112213850021362, |
| "reward_std": 0.30633312463760376, |
| "rewards/GDino": 0.7401995956897736, |
| "rewards/GIT": 0.30288365483283997, |
| "rewards/HPSv2": 0.2552833557128906, |
| "rewards/ORM": 0.6128547042608261, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.125, |
| "step": 38 |
| }, |
| { |
| "completion_length": 67.6875, |
| "epoch": 0.04318936877076412, |
| "grad_norm": 0.6357575058937073, |
| "kl": 0.01482391357421875, |
| "learning_rate": 9.756249999999999e-07, |
| "loss": 0.023865018505603075, |
| "reward": 2.345404624938965, |
| "reward_std": 0.31367097795009613, |
| "rewards/GDino": 0.8703815042972565, |
| "rewards/GIT": 0.4902418553829193, |
| "rewards/HPSv2": 0.26603126525878906, |
| "rewards/ORM": 0.71875, |
| "self_certainty_semantic": -25.5625, |
| "self_certainty_token": -21.75, |
| "step": 39 |
| }, |
| { |
| "completion_length": 59.671875, |
| "epoch": 0.044296788482835, |
| "grad_norm": 0.5422465801239014, |
| "kl": 0.00281524658203125, |
| "learning_rate": 9.75e-07, |
| "loss": -0.018710695207118988, |
| "reward": 2.222834825515747, |
| "reward_std": 0.42842796444892883, |
| "rewards/GDino": 0.8634105622768402, |
| "rewards/GIT": 0.40908148139715195, |
| "rewards/HPSv2": 0.27498817443847656, |
| "rewards/ORM": 0.6753546893596649, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.125, |
| "step": 40 |
| }, |
| { |
| "completion_length": 60.9375, |
| "epoch": 0.04540420819490587, |
| "grad_norm": 0.7511593103408813, |
| "kl": 0.00299072265625, |
| "learning_rate": 9.743749999999999e-07, |
| "loss": 0.005782268475741148, |
| "reward": 1.8980144262313843, |
| "reward_std": 0.3208035007119179, |
| "rewards/GDino": 0.6784752607345581, |
| "rewards/GIT": 0.3914954513311386, |
| "rewards/HPSv2": 0.24643898010253906, |
| "rewards/ORM": 0.5816046595573425, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.25, |
| "step": 41 |
| }, |
| { |
| "completion_length": 48.4375, |
| "epoch": 0.046511627906976744, |
| "grad_norm": 0.5177002549171448, |
| "kl": 0.0025177001953125, |
| "learning_rate": 9.7375e-07, |
| "loss": 0.045526545494794846, |
| "reward": 2.269711136817932, |
| "reward_std": 0.48014624416828156, |
| "rewards/GDino": 0.8855312466621399, |
| "rewards/GIT": 0.4437972754240036, |
| "rewards/HPSv2": 0.2572154998779297, |
| "rewards/ORM": 0.6831671893596649, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -20.5625, |
| "step": 42 |
| }, |
| { |
| "completion_length": 67.875, |
| "epoch": 0.047619047619047616, |
| "grad_norm": 0.5885121822357178, |
| "kl": 0.002044677734375, |
| "learning_rate": 9.73125e-07, |
| "loss": 0.013573684729635715, |
| "reward": 1.6382005214691162, |
| "reward_std": 0.38919302821159363, |
| "rewards/GDino": 0.6114583313465118, |
| "rewards/GIT": 0.3806646466255188, |
| "rewards/HPSv2": 0.23286819458007812, |
| "rewards/ORM": 0.41320937871932983, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.0625, |
| "step": 43 |
| }, |
| { |
| "completion_length": 54.4375, |
| "epoch": 0.048726467331118496, |
| "grad_norm": 0.40727919340133667, |
| "kl": 0.0020751953125, |
| "learning_rate": 9.725e-07, |
| "loss": -0.01244093757122755, |
| "reward": 2.8831005096435547, |
| "reward_std": 0.31665875762701035, |
| "rewards/GDino": 0.9588541388511658, |
| "rewards/GIT": 0.7738310992717743, |
| "rewards/HPSv2": 0.2601909637451172, |
| "rewards/ORM": 0.8902243673801422, |
| "self_certainty_semantic": -25.5625, |
| "self_certainty_token": -21.0625, |
| "step": 44 |
| }, |
| { |
| "completion_length": 54.90625, |
| "epoch": 0.04983388704318937, |
| "grad_norm": 0.4928445816040039, |
| "kl": 0.0024566650390625, |
| "learning_rate": 9.71875e-07, |
| "loss": 0.00010553281754255295, |
| "reward": 2.4343937635421753, |
| "reward_std": 0.5984751731157303, |
| "rewards/GDino": 0.862500011920929, |
| "rewards/GIT": 0.5139474421739578, |
| "rewards/HPSv2": 0.26379966735839844, |
| "rewards/ORM": 0.7941466569900513, |
| "self_certainty_semantic": -25.5625, |
| "self_certainty_token": -21.25, |
| "step": 45 |
| }, |
| { |
| "completion_length": 56.078125, |
| "epoch": 0.05094130675526024, |
| "grad_norm": 0.37051326036453247, |
| "kl": 0.00231170654296875, |
| "learning_rate": 9.712499999999998e-07, |
| "loss": 0.007893505971878767, |
| "reward": 1.9575175046920776, |
| "reward_std": 0.3945648521184921, |
| "rewards/GDino": 0.5999999940395355, |
| "rewards/GIT": 0.32395021617412567, |
| "rewards/HPSv2": 0.26719093322753906, |
| "rewards/ORM": 0.7663763463497162, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0625, |
| "step": 46 |
| }, |
| { |
| "completion_length": 55.171875, |
| "epoch": 0.05204872646733112, |
| "grad_norm": 0.8945181369781494, |
| "kl": 0.0025634765625, |
| "learning_rate": 9.70625e-07, |
| "loss": -0.0013387980870902538, |
| "reward": 1.836871862411499, |
| "reward_std": 0.23468619585037231, |
| "rewards/GDino": 0.7209739089012146, |
| "rewards/GIT": 0.22856376320123672, |
| "rewards/HPSv2": 0.27921295166015625, |
| "rewards/ORM": 0.608121246099472, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.6875, |
| "step": 47 |
| }, |
| { |
| "completion_length": 57.984375, |
| "epoch": 0.053156146179401995, |
| "grad_norm": 1.6689982414245605, |
| "kl": 0.00267791748046875, |
| "learning_rate": 9.7e-07, |
| "loss": 0.022647732868790627, |
| "reward": 1.454766035079956, |
| "reward_std": 0.40884387493133545, |
| "rewards/GDino": 0.6050891876220703, |
| "rewards/GIT": 0.0, |
| "rewards/HPSv2": 0.2698974609375, |
| "rewards/ORM": 0.5797793865203857, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.625, |
| "step": 48 |
| }, |
| { |
| "completion_length": 58.046875, |
| "epoch": 0.05426356589147287, |
| "grad_norm": 0.4761441648006439, |
| "kl": 0.002048492431640625, |
| "learning_rate": 9.69375e-07, |
| "loss": 0.016307475278154016, |
| "reward": 1.9066129326820374, |
| "reward_std": 0.5319462567567825, |
| "rewards/GDino": 0.7744874656200409, |
| "rewards/GIT": 0.2370736114680767, |
| "rewards/HPSv2": 0.2514495849609375, |
| "rewards/ORM": 0.6436022371053696, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.625, |
| "step": 49 |
| }, |
| { |
| "completion_length": 61.0, |
| "epoch": 0.05537098560354374, |
| "grad_norm": 0.8074173331260681, |
| "kl": 0.0040283203125, |
| "learning_rate": 9.6875e-07, |
| "loss": 0.005913220578804612, |
| "reward": 2.0915883779525757, |
| "reward_std": 0.5395111739635468, |
| "rewards/GDino": 0.7859093248844147, |
| "rewards/GIT": 0.3929952085018158, |
| "rewards/HPSv2": 0.25482940673828125, |
| "rewards/ORM": 0.657854437828064, |
| "self_certainty_semantic": -25.5625, |
| "self_certainty_token": -22.25, |
| "step": 50 |
| }, |
| { |
| "completion_length": 44.359375, |
| "epoch": 0.05647840531561462, |
| "grad_norm": 0.5618427991867065, |
| "kl": 0.002471923828125, |
| "learning_rate": 9.68125e-07, |
| "loss": -0.003945098840631545, |
| "reward": 1.8058671951293945, |
| "reward_std": 0.5712144523859024, |
| "rewards/GDino": 0.7815796732902527, |
| "rewards/GIT": 0.2604931816458702, |
| "rewards/HPSv2": 0.27115440368652344, |
| "rewards/ORM": 0.49263995885849, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -22.625, |
| "step": 51 |
| }, |
| { |
| "completion_length": 48.0, |
| "epoch": 0.05758582502768549, |
| "grad_norm": 107.57159423828125, |
| "kl": 26.37615966796875, |
| "learning_rate": 9.675e-07, |
| "loss": 0.27801212295889854, |
| "reward": 2.4165316820144653, |
| "reward_std": 0.2998274937272072, |
| "rewards/GDino": 0.9244791567325592, |
| "rewards/GIT": 0.6574473828077316, |
| "rewards/HPSv2": 0.2756366729736328, |
| "rewards/ORM": 0.5589684545993805, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.375, |
| "step": 52 |
| }, |
| { |
| "completion_length": 52.140625, |
| "epoch": 0.058693244739756366, |
| "grad_norm": 0.4408358931541443, |
| "kl": 0.00232696533203125, |
| "learning_rate": 9.66875e-07, |
| "loss": 0.013528472045436502, |
| "reward": 1.8899905681610107, |
| "reward_std": 0.4558149725198746, |
| "rewards/GDino": 0.730059951543808, |
| "rewards/GIT": 0.39098620414733887, |
| "rewards/HPSv2": 0.24242782592773438, |
| "rewards/ORM": 0.5265165567398071, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -20.5625, |
| "step": 53 |
| }, |
| { |
| "completion_length": 58.390625, |
| "epoch": 0.059800664451827246, |
| "grad_norm": 0.48384228348731995, |
| "kl": 0.00225067138671875, |
| "learning_rate": 9.6625e-07, |
| "loss": 0.005568797350861132, |
| "reward": 1.638724684715271, |
| "reward_std": 0.41337575018405914, |
| "rewards/GDino": 0.6137361526489258, |
| "rewards/GIT": 0.24863167852163315, |
| "rewards/HPSv2": 0.24831581115722656, |
| "rewards/ORM": 0.5280410945415497, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -20.9375, |
| "step": 54 |
| }, |
| { |
| "completion_length": 50.234375, |
| "epoch": 0.06090808416389812, |
| "grad_norm": 0.46963369846343994, |
| "kl": 0.0026397705078125, |
| "learning_rate": 9.65625e-07, |
| "loss": 0.009267964400351048, |
| "reward": 1.7191376686096191, |
| "reward_std": 0.521537572145462, |
| "rewards/GDino": 0.7086881995201111, |
| "rewards/GIT": 0.3270767852663994, |
| "rewards/HPSv2": 0.2678356170654297, |
| "rewards/ORM": 0.4155370891094208, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -21.6875, |
| "step": 55 |
| }, |
| { |
| "completion_length": 59.953125, |
| "epoch": 0.06201550387596899, |
| "grad_norm": 0.6913841366767883, |
| "kl": 0.0024261474609375, |
| "learning_rate": 9.649999999999999e-07, |
| "loss": 0.03414425998926163, |
| "reward": 1.9336698055267334, |
| "reward_std": 0.45749759674072266, |
| "rewards/GDino": 0.6963726580142975, |
| "rewards/GIT": 0.38425514101982117, |
| "rewards/HPSv2": 0.2471466064453125, |
| "rewards/ORM": 0.6058953106403351, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.875, |
| "step": 56 |
| }, |
| { |
| "completion_length": 50.765625, |
| "epoch": 0.06312292358803986, |
| "grad_norm": 0.5066769123077393, |
| "kl": 0.002532958984375, |
| "learning_rate": 9.64375e-07, |
| "loss": 0.009842937346547842, |
| "reward": 1.8338811993598938, |
| "reward_std": 0.3951306492090225, |
| "rewards/GDino": 0.7909577786922455, |
| "rewards/GIT": 0.24781160056591034, |
| "rewards/HPSv2": 0.2509651184082031, |
| "rewards/ORM": 0.5441466569900513, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -21.4375, |
| "step": 57 |
| }, |
| { |
| "completion_length": 52.9375, |
| "epoch": 0.06423034330011074, |
| "grad_norm": 0.37791869044303894, |
| "kl": 0.002685546875, |
| "learning_rate": 9.637499999999999e-07, |
| "loss": 0.024126023054122925, |
| "reward": 1.8852884769439697, |
| "reward_std": 0.46756890416145325, |
| "rewards/GDino": 0.732811689376831, |
| "rewards/GIT": 0.38145140558481216, |
| "rewards/HPSv2": 0.2541465759277344, |
| "rewards/ORM": 0.5168787688016891, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.25, |
| "step": 58 |
| }, |
| { |
| "completion_length": 47.234375, |
| "epoch": 0.06533776301218161, |
| "grad_norm": 0.7410405278205872, |
| "kl": 0.0026092529296875, |
| "learning_rate": 9.63125e-07, |
| "loss": -0.01674468442797661, |
| "reward": 2.3462648391723633, |
| "reward_std": 0.2433818019926548, |
| "rewards/GDino": 0.8425607979297638, |
| "rewards/GIT": 0.46571947634220123, |
| "rewards/HPSv2": 0.2664222717285156, |
| "rewards/ORM": 0.771562248468399, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.25, |
| "step": 59 |
| }, |
| { |
| "completion_length": 45.015625, |
| "epoch": 0.0664451827242525, |
| "grad_norm": 0.5326105952262878, |
| "kl": 0.0026397705078125, |
| "learning_rate": 9.624999999999999e-07, |
| "loss": 0.003804182168096304, |
| "reward": 2.036432147026062, |
| "reward_std": 0.3990803211927414, |
| "rewards/GDino": 0.8798050284385681, |
| "rewards/GIT": 0.4744318723678589, |
| "rewards/HPSv2": 0.238006591796875, |
| "rewards/ORM": 0.44418865442276, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -20.6875, |
| "step": 60 |
| }, |
| { |
| "completion_length": 65.0, |
| "epoch": 0.06755260243632337, |
| "grad_norm": 0.5713196396827698, |
| "kl": 0.00235748291015625, |
| "learning_rate": 9.61875e-07, |
| "loss": 0.04368375800549984, |
| "reward": 2.1398236751556396, |
| "reward_std": 0.3530130609869957, |
| "rewards/GDino": 0.7138020694255829, |
| "rewards/GIT": 0.644903838634491, |
| "rewards/HPSv2": 0.2529468536376953, |
| "rewards/ORM": 0.5281709432601929, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -20.6875, |
| "step": 61 |
| }, |
| { |
| "completion_length": 54.9375, |
| "epoch": 0.06866002214839424, |
| "grad_norm": 5.612445831298828, |
| "kl": 0.00339508056640625, |
| "learning_rate": 9.6125e-07, |
| "loss": 0.008875304833054543, |
| "reward": 2.497900605201721, |
| "reward_std": 0.41675496101379395, |
| "rewards/GDino": 0.872697502374649, |
| "rewards/GIT": 0.601748138666153, |
| "rewards/HPSv2": 0.2640380859375, |
| "rewards/ORM": 0.759416937828064, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -20.875, |
| "step": 62 |
| }, |
| { |
| "completion_length": 49.734375, |
| "epoch": 0.06976744186046512, |
| "grad_norm": 0.5861217379570007, |
| "kl": 0.003326416015625, |
| "learning_rate": 9.606249999999998e-07, |
| "loss": 0.01025251136161387, |
| "reward": 2.2640050053596497, |
| "reward_std": 0.48744213581085205, |
| "rewards/GDino": 0.8172852694988251, |
| "rewards/GIT": 0.44742196798324585, |
| "rewards/HPSv2": 0.2430896759033203, |
| "rewards/ORM": 0.7562080323696136, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.1875, |
| "step": 63 |
| }, |
| { |
| "completion_length": 64.375, |
| "epoch": 0.07087486157253599, |
| "grad_norm": 0.39266109466552734, |
| "kl": 0.00298309326171875, |
| "learning_rate": 9.6e-07, |
| "loss": -0.005469436291605234, |
| "reward": 1.6910768747329712, |
| "reward_std": 0.2151722088456154, |
| "rewards/GDino": 0.7097718715667725, |
| "rewards/GIT": 0.32366037368774414, |
| "rewards/HPSv2": 0.2576026916503906, |
| "rewards/ORM": 0.40004195272922516, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -21.9375, |
| "step": 64 |
| }, |
| { |
| "completion_length": 61.515625, |
| "epoch": 0.07198228128460686, |
| "grad_norm": 0.705937922000885, |
| "kl": 0.002685546875, |
| "learning_rate": 9.59375e-07, |
| "loss": 0.010601098649203777, |
| "reward": 2.128853142261505, |
| "reward_std": 0.4351096749305725, |
| "rewards/GDino": 0.7197916805744171, |
| "rewards/GIT": 0.6168824732303619, |
| "rewards/HPSv2": 0.23163414001464844, |
| "rewards/ORM": 0.5605448335409164, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.75, |
| "step": 65 |
| }, |
| { |
| "completion_length": 49.0, |
| "epoch": 0.07308970099667775, |
| "grad_norm": 0.4427480101585388, |
| "kl": 0.002899169921875, |
| "learning_rate": 9.5875e-07, |
| "loss": 0.02646360918879509, |
| "reward": 2.1654986143112183, |
| "reward_std": 0.37753987312316895, |
| "rewards/GDino": 0.6895833611488342, |
| "rewards/GIT": 0.48387444019317627, |
| "rewards/HPSv2": 0.2579364776611328, |
| "rewards/ORM": 0.7341042160987854, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.5625, |
| "step": 66 |
| }, |
| { |
| "completion_length": 63.140625, |
| "epoch": 0.07419712070874862, |
| "grad_norm": 0.7619237899780273, |
| "kl": 0.00284576416015625, |
| "learning_rate": 9.58125e-07, |
| "loss": 0.026691121514886618, |
| "reward": 2.3450592160224915, |
| "reward_std": 0.2740027904510498, |
| "rewards/GDino": 0.8025760054588318, |
| "rewards/GIT": 0.5677543580532074, |
| "rewards/HPSv2": 0.2594585418701172, |
| "rewards/ORM": 0.7152703106403351, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.875, |
| "step": 67 |
| }, |
| { |
| "completion_length": 50.171875, |
| "epoch": 0.0753045404208195, |
| "grad_norm": 0.4760603904724121, |
| "kl": 0.0030517578125, |
| "learning_rate": 9.575e-07, |
| "loss": 0.022392848506569862, |
| "reward": 1.6361079216003418, |
| "reward_std": 0.33574268221855164, |
| "rewards/GDino": 0.6061920523643494, |
| "rewards/GIT": 0.31722745299339294, |
| "rewards/HPSv2": 0.2595634460449219, |
| "rewards/ORM": 0.453125, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.5, |
| "step": 68 |
| }, |
| { |
| "completion_length": 55.046875, |
| "epoch": 0.07641196013289037, |
| "grad_norm": 0.5907943248748779, |
| "kl": 0.00336456298828125, |
| "learning_rate": 9.56875e-07, |
| "loss": -0.0030646873638033867, |
| "reward": 2.119426429271698, |
| "reward_std": 0.298831045627594, |
| "rewards/GDino": 0.8028125166893005, |
| "rewards/GIT": 0.3893257826566696, |
| "rewards/HPSv2": 0.2710380554199219, |
| "rewards/ORM": 0.65625, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.4375, |
| "step": 69 |
| }, |
| { |
| "completion_length": 51.140625, |
| "epoch": 0.07751937984496124, |
| "grad_norm": 0.47751373052597046, |
| "kl": 0.00359344482421875, |
| "learning_rate": 9.5625e-07, |
| "loss": -0.011344656813889742, |
| "reward": 1.4646188020706177, |
| "reward_std": 0.5817874372005463, |
| "rewards/GDino": 0.5935695767402649, |
| "rewards/GIT": 0.23897356167435646, |
| "rewards/HPSv2": 0.25234222412109375, |
| "rewards/ORM": 0.37973344326019287, |
| "self_certainty_semantic": -24.875, |
| "self_certainty_token": -20.9375, |
| "step": 70 |
| }, |
| { |
| "completion_length": 55.609375, |
| "epoch": 0.07862679955703211, |
| "grad_norm": 0.5281980633735657, |
| "kl": 0.0030364990234375, |
| "learning_rate": 9.556249999999999e-07, |
| "loss": -0.023217559792101383, |
| "reward": 1.856022596359253, |
| "reward_std": 0.4435942769050598, |
| "rewards/GDino": 0.6947268545627594, |
| "rewards/GIT": 0.28702250868082047, |
| "rewards/HPSv2": 0.26489830017089844, |
| "rewards/ORM": 0.609375, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -20.5, |
| "step": 71 |
| }, |
| { |
| "completion_length": 42.390625, |
| "epoch": 0.07973421926910298, |
| "grad_norm": 0.4538242518901825, |
| "kl": 0.002960205078125, |
| "learning_rate": 9.55e-07, |
| "loss": 0.016265914775431156, |
| "reward": 1.911847174167633, |
| "reward_std": 0.4016146659851074, |
| "rewards/GDino": 0.6731474995613098, |
| "rewards/GIT": 0.46439771354198456, |
| "rewards/HPSv2": 0.2497406005859375, |
| "rewards/ORM": 0.524561420083046, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -20.875, |
| "step": 72 |
| }, |
| { |
| "completion_length": 53.28125, |
| "epoch": 0.08084163898117387, |
| "grad_norm": 0.5773823261260986, |
| "kl": 0.00330352783203125, |
| "learning_rate": 9.543749999999999e-07, |
| "loss": -0.0016377167776226997, |
| "reward": 2.114488363265991, |
| "reward_std": 0.44427454471588135, |
| "rewards/GDino": 0.8240922689437866, |
| "rewards/GIT": 0.4950668513774872, |
| "rewards/HPSv2": 0.24412155151367188, |
| "rewards/ORM": 0.5512077808380127, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.0, |
| "step": 73 |
| }, |
| { |
| "completion_length": 56.296875, |
| "epoch": 0.08194905869324474, |
| "grad_norm": 0.43449509143829346, |
| "kl": 0.0035247802734375, |
| "learning_rate": 9.5375e-07, |
| "loss": 0.03005522396415472, |
| "reward": 2.32301664352417, |
| "reward_std": 0.22542773187160492, |
| "rewards/GDino": 0.864062488079071, |
| "rewards/GIT": 0.5282620340585709, |
| "rewards/HPSv2": 0.25408363342285156, |
| "rewards/ORM": 0.6766084730625153, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.25, |
| "step": 74 |
| }, |
| { |
| "completion_length": 67.6875, |
| "epoch": 0.08305647840531562, |
| "grad_norm": 0.4218258857727051, |
| "kl": 0.0028228759765625, |
| "learning_rate": 9.53125e-07, |
| "loss": 0.015081442426890135, |
| "reward": 1.7625158429145813, |
| "reward_std": 0.4334114193916321, |
| "rewards/GDino": 0.6663236618041992, |
| "rewards/GIT": 0.26877461373806, |
| "rewards/HPSv2": 0.2647876739501953, |
| "rewards/ORM": 0.5626298785209656, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.125, |
| "step": 75 |
| }, |
| { |
| "completion_length": 62.15625, |
| "epoch": 0.08416389811738649, |
| "grad_norm": 0.45278123021125793, |
| "kl": 0.00312042236328125, |
| "learning_rate": 9.525e-07, |
| "loss": 0.01650754688307643, |
| "reward": 2.2938578128814697, |
| "reward_std": 0.5077499151229858, |
| "rewards/GDino": 0.7734375, |
| "rewards/GIT": 0.6401466727256775, |
| "rewards/HPSv2": 0.2568778991699219, |
| "rewards/ORM": 0.6233955323696136, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.125, |
| "step": 76 |
| }, |
| { |
| "completion_length": 50.984375, |
| "epoch": 0.08527131782945736, |
| "grad_norm": 0.5513558387756348, |
| "kl": 0.004730224609375, |
| "learning_rate": 9.51875e-07, |
| "loss": -0.008258584188297391, |
| "reward": 1.6354877948760986, |
| "reward_std": 0.5420883148908615, |
| "rewards/GDino": 0.643737405538559, |
| "rewards/GIT": 0.20579323172569275, |
| "rewards/HPSv2": 0.2405567169189453, |
| "rewards/ORM": 0.5454003810882568, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.1875, |
| "step": 77 |
| }, |
| { |
| "completion_length": 56.390625, |
| "epoch": 0.08637873754152824, |
| "grad_norm": 0.9578920602798462, |
| "kl": 0.00360107421875, |
| "learning_rate": 9.5125e-07, |
| "loss": 0.0016261041164398193, |
| "reward": 2.061507523059845, |
| "reward_std": 0.2758500352501869, |
| "rewards/GDino": 0.7561410367488861, |
| "rewards/GIT": 0.33666322380304337, |
| "rewards/HPSv2": 0.2762489318847656, |
| "rewards/ORM": 0.6924542784690857, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.1875, |
| "step": 78 |
| }, |
| { |
| "completion_length": 57.375, |
| "epoch": 0.08748615725359911, |
| "grad_norm": 0.46459418535232544, |
| "kl": 0.004241943359375, |
| "learning_rate": 9.50625e-07, |
| "loss": -0.019409675151109695, |
| "reward": 2.298323154449463, |
| "reward_std": 0.22066934406757355, |
| "rewards/GDino": 0.8136925399303436, |
| "rewards/GIT": 0.6333461850881577, |
| "rewards/HPSv2": 0.27008056640625, |
| "rewards/ORM": 0.5812040567398071, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.875, |
| "step": 79 |
| }, |
| { |
| "completion_length": 60.5625, |
| "epoch": 0.08859357696567, |
| "grad_norm": 0.4274587631225586, |
| "kl": 0.004058837890625, |
| "learning_rate": 9.499999999999999e-07, |
| "loss": 0.013256619684398174, |
| "reward": 1.6786987781524658, |
| "reward_std": 0.3984425514936447, |
| "rewards/GDino": 0.6007516384124756, |
| "rewards/GIT": 0.18326736986637115, |
| "rewards/HPSv2": 0.2720355987548828, |
| "rewards/ORM": 0.6226442158222198, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -21.3125, |
| "step": 80 |
| }, |
| { |
| "completion_length": 58.03125, |
| "epoch": 0.08970099667774087, |
| "grad_norm": 0.9172859191894531, |
| "kl": 0.00426483154296875, |
| "learning_rate": 9.493749999999999e-07, |
| "loss": 0.003496276680380106, |
| "reward": 2.106017231941223, |
| "reward_std": 0.30050399899482727, |
| "rewards/GDino": 0.7440759837627411, |
| "rewards/GIT": 0.3581302911043167, |
| "rewards/HPSv2": 0.27126121520996094, |
| "rewards/ORM": 0.7325496971607208, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -20.875, |
| "step": 81 |
| }, |
| { |
| "completion_length": 49.5625, |
| "epoch": 0.09080841638981174, |
| "grad_norm": 0.4841405153274536, |
| "kl": 0.00412750244140625, |
| "learning_rate": 9.487499999999999e-07, |
| "loss": 0.025506282225251198, |
| "reward": 1.6879253387451172, |
| "reward_std": 0.42353254556655884, |
| "rewards/GDino": 0.6098452508449554, |
| "rewards/GIT": 0.38033944368362427, |
| "rewards/HPSv2": 0.2658271789550781, |
| "rewards/ORM": 0.43191343545913696, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -20.9375, |
| "step": 82 |
| }, |
| { |
| "completion_length": 48.328125, |
| "epoch": 0.09191583610188261, |
| "grad_norm": 0.492243230342865, |
| "kl": 0.00345611572265625, |
| "learning_rate": 9.481249999999999e-07, |
| "loss": -0.0034960508346557617, |
| "reward": 2.1111596822738647, |
| "reward_std": 0.41540510952472687, |
| "rewards/GDino": 0.7717877924442291, |
| "rewards/GIT": 0.4860316216945648, |
| "rewards/HPSv2": 0.2670021057128906, |
| "rewards/ORM": 0.5863381326198578, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.4375, |
| "step": 83 |
| }, |
| { |
| "completion_length": 66.3125, |
| "epoch": 0.09302325581395349, |
| "grad_norm": 0.5617808699607849, |
| "kl": 0.004180908203125, |
| "learning_rate": 9.474999999999999e-07, |
| "loss": 0.003248518332839012, |
| "reward": 2.094790816307068, |
| "reward_std": 0.3879907354712486, |
| "rewards/GDino": 0.7973622679710388, |
| "rewards/GIT": 0.632976621389389, |
| "rewards/HPSv2": 0.24137306213378906, |
| "rewards/ORM": 0.4230788052082062, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.1875, |
| "step": 84 |
| }, |
| { |
| "completion_length": 51.40625, |
| "epoch": 0.09413067552602436, |
| "grad_norm": 0.5695884823799133, |
| "kl": 0.003204345703125, |
| "learning_rate": 9.468749999999999e-07, |
| "loss": 0.012543351389467716, |
| "reward": 1.8675293326377869, |
| "reward_std": 0.4282868355512619, |
| "rewards/GDino": 0.6550000011920929, |
| "rewards/GIT": 0.33260630816221237, |
| "rewards/HPSv2": 0.24515533447265625, |
| "rewards/ORM": 0.6347676515579224, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.375, |
| "step": 85 |
| }, |
| { |
| "completion_length": 48.296875, |
| "epoch": 0.09523809523809523, |
| "grad_norm": 0.46590158343315125, |
| "kl": 0.00469970703125, |
| "learning_rate": 9.462499999999999e-07, |
| "loss": 0.00347991194576025, |
| "reward": 2.2731298208236694, |
| "reward_std": 0.383390873670578, |
| "rewards/GDino": 0.8246111273765564, |
| "rewards/GIT": 0.33447980135679245, |
| "rewards/HPSv2": 0.29212188720703125, |
| "rewards/ORM": 0.821916937828064, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.0, |
| "step": 86 |
| }, |
| { |
| "completion_length": 54.390625, |
| "epoch": 0.09634551495016612, |
| "grad_norm": 0.5397853255271912, |
| "kl": 0.004425048828125, |
| "learning_rate": 9.45625e-07, |
| "loss": 0.008617566898465157, |
| "reward": 2.2459940314292908, |
| "reward_std": 0.4676859378814697, |
| "rewards/GDino": 0.7356771230697632, |
| "rewards/GIT": 0.46453191339969635, |
| "rewards/HPSv2": 0.26766395568847656, |
| "rewards/ORM": 0.7781210243701935, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.0, |
| "step": 87 |
| }, |
| { |
| "completion_length": 42.15625, |
| "epoch": 0.09745293466223699, |
| "grad_norm": 0.48280662298202515, |
| "kl": 0.00406646728515625, |
| "learning_rate": 9.45e-07, |
| "loss": 0.016791983507573605, |
| "reward": 2.1528985500335693, |
| "reward_std": 0.44025059044361115, |
| "rewards/GDino": 0.7985424101352692, |
| "rewards/GIT": 0.47699007391929626, |
| "rewards/HPSv2": 0.2789325714111328, |
| "rewards/ORM": 0.5984334945678711, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -20.875, |
| "step": 88 |
| }, |
| { |
| "completion_length": 48.25, |
| "epoch": 0.09856035437430787, |
| "grad_norm": 0.4512772560119629, |
| "kl": 0.00405120849609375, |
| "learning_rate": 9.44375e-07, |
| "loss": -0.009609811007976532, |
| "reward": 2.155352771282196, |
| "reward_std": 0.3193782642483711, |
| "rewards/GDino": 0.7525902688503265, |
| "rewards/GIT": 0.4481022357940674, |
| "rewards/HPSv2": 0.2619743347167969, |
| "rewards/ORM": 0.6926859021186829, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.125, |
| "step": 89 |
| }, |
| { |
| "completion_length": 47.78125, |
| "epoch": 0.09966777408637874, |
| "grad_norm": 0.5204576849937439, |
| "kl": 0.004425048828125, |
| "learning_rate": 9.4375e-07, |
| "loss": -0.017570611089468002, |
| "reward": 2.3318194150924683, |
| "reward_std": 0.3641355484724045, |
| "rewards/GDino": 0.854687511920929, |
| "rewards/GIT": 0.6271218061447144, |
| "rewards/HPSv2": 0.26286888122558594, |
| "rewards/ORM": 0.5871412754058838, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -21.4375, |
| "step": 90 |
| }, |
| { |
| "completion_length": 54.9375, |
| "epoch": 0.10077519379844961, |
| "grad_norm": 0.7515896558761597, |
| "kl": 0.0042266845703125, |
| "learning_rate": 9.43125e-07, |
| "loss": 0.022024651989340782, |
| "reward": 1.7255874276161194, |
| "reward_std": 0.3924099802970886, |
| "rewards/GDino": 0.6800954043865204, |
| "rewards/GIT": 0.41760827600955963, |
| "rewards/HPSv2": 0.22957611083984375, |
| "rewards/ORM": 0.3983076214790344, |
| "self_certainty_semantic": -24.875, |
| "self_certainty_token": -21.125, |
| "step": 91 |
| }, |
| { |
| "completion_length": 51.90625, |
| "epoch": 0.10188261351052048, |
| "grad_norm": 0.6844750046730042, |
| "kl": 0.004730224609375, |
| "learning_rate": 9.425e-07, |
| "loss": 0.017017286270856857, |
| "reward": 1.7472361326217651, |
| "reward_std": 0.49342362582683563, |
| "rewards/GDino": 0.7615998685359955, |
| "rewards/GIT": 0.3799494504928589, |
| "rewards/HPSv2": 0.2450580596923828, |
| "rewards/ORM": 0.36062875390052795, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -21.0625, |
| "step": 92 |
| }, |
| { |
| "completion_length": 52.5625, |
| "epoch": 0.10299003322259136, |
| "grad_norm": 0.476144403219223, |
| "kl": 0.004547119140625, |
| "learning_rate": 9.41875e-07, |
| "loss": -0.006627652794122696, |
| "reward": 2.3529324531555176, |
| "reward_std": 0.38789400458335876, |
| "rewards/GDino": 0.8122400343418121, |
| "rewards/GIT": 0.40920257568359375, |
| "rewards/HPSv2": 0.25894737243652344, |
| "rewards/ORM": 0.8725424408912659, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -20.75, |
| "step": 93 |
| }, |
| { |
| "completion_length": 45.234375, |
| "epoch": 0.10409745293466224, |
| "grad_norm": 0.4303518235683441, |
| "kl": 0.00390625, |
| "learning_rate": 9.4125e-07, |
| "loss": 0.002329372800886631, |
| "reward": 2.063507556915283, |
| "reward_std": 0.4875355362892151, |
| "rewards/GDino": 0.8157378733158112, |
| "rewards/GIT": 0.2162991166114807, |
| "rewards/HPSv2": 0.2860240936279297, |
| "rewards/ORM": 0.7454463839530945, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.25, |
| "step": 94 |
| }, |
| { |
| "completion_length": 54.21875, |
| "epoch": 0.10520487264673312, |
| "grad_norm": 0.9745371341705322, |
| "kl": 0.004852294921875, |
| "learning_rate": 9.40625e-07, |
| "loss": 0.015892890747636557, |
| "reward": 2.4900766611099243, |
| "reward_std": 0.33158986270427704, |
| "rewards/GDino": 0.9456690549850464, |
| "rewards/GIT": 0.7110534906387329, |
| "rewards/HPSv2": 0.2568836212158203, |
| "rewards/ORM": 0.5764705836772919, |
| "self_certainty_semantic": -25.625, |
| "self_certainty_token": -21.3125, |
| "step": 95 |
| }, |
| { |
| "completion_length": 62.953125, |
| "epoch": 0.10631229235880399, |
| "grad_norm": 1.6108874082565308, |
| "kl": 0.00475311279296875, |
| "learning_rate": 9.399999999999999e-07, |
| "loss": 0.012537557166069746, |
| "reward": 2.4274561405181885, |
| "reward_std": 0.3028244078159332, |
| "rewards/GDino": 0.9155160486698151, |
| "rewards/GIT": 0.6933247745037079, |
| "rewards/HPSv2": 0.25919437408447266, |
| "rewards/ORM": 0.5594209432601929, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -20.3125, |
| "step": 96 |
| }, |
| { |
| "completion_length": 44.890625, |
| "epoch": 0.10741971207087486, |
| "grad_norm": 0.42777886986732483, |
| "kl": 0.0064697265625, |
| "learning_rate": 9.393749999999999e-07, |
| "loss": 0.006582918576896191, |
| "reward": 1.7229499220848083, |
| "reward_std": 0.29571742564439774, |
| "rewards/GDino": 0.6976552903652191, |
| "rewards/GIT": 0.17514611035585403, |
| "rewards/HPSv2": 0.2757740020751953, |
| "rewards/ORM": 0.5743745565414429, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.5, |
| "step": 97 |
| }, |
| { |
| "completion_length": 52.78125, |
| "epoch": 0.10852713178294573, |
| "grad_norm": 0.4346785247325897, |
| "kl": 0.00446319580078125, |
| "learning_rate": 9.387499999999999e-07, |
| "loss": 0.010664775501936674, |
| "reward": 1.9896260499954224, |
| "reward_std": 0.5384568274021149, |
| "rewards/GDino": 0.7534899115562439, |
| "rewards/GIT": 0.416723370552063, |
| "rewards/HPSv2": 0.25490760803222656, |
| "rewards/ORM": 0.5645051300525665, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.4375, |
| "step": 98 |
| }, |
| { |
| "completion_length": 53.546875, |
| "epoch": 0.10963455149501661, |
| "grad_norm": 0.4226502478122711, |
| "kl": 0.00543212890625, |
| "learning_rate": 9.381249999999999e-07, |
| "loss": -0.009754271944984794, |
| "reward": 2.1711018085479736, |
| "reward_std": 0.3036491945385933, |
| "rewards/GDino": 0.8239583373069763, |
| "rewards/GIT": 0.6844146698713303, |
| "rewards/HPSv2": 0.24951934814453125, |
| "rewards/ORM": 0.41320937871932983, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.0, |
| "step": 99 |
| }, |
| { |
| "completion_length": 48.0625, |
| "epoch": 0.11074197120708748, |
| "grad_norm": 0.4250389039516449, |
| "kl": 0.00537109375, |
| "learning_rate": 9.374999999999999e-07, |
| "loss": -0.014408082235604525, |
| "reward": 1.9375371932983398, |
| "reward_std": 0.4484590142965317, |
| "rewards/GDino": 0.6897697150707245, |
| "rewards/GIT": 0.4094943553209305, |
| "rewards/HPSv2": 0.2472515106201172, |
| "rewards/ORM": 0.5910216569900513, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.4375, |
| "step": 100 |
| }, |
| { |
| "completion_length": 53.546875, |
| "epoch": 0.11184939091915837, |
| "grad_norm": 0.9339170455932617, |
| "kl": 0.00566864013671875, |
| "learning_rate": 9.368749999999999e-07, |
| "loss": 0.003982411697506905, |
| "reward": 2.2582755088806152, |
| "reward_std": 0.41422703862190247, |
| "rewards/GDino": 0.8809943795204163, |
| "rewards/GIT": 0.5047063827514648, |
| "rewards/HPSv2": 0.27569580078125, |
| "rewards/ORM": 0.5968790352344513, |
| "self_certainty_semantic": -25.6875, |
| "self_certainty_token": -20.6875, |
| "step": 101 |
| }, |
| { |
| "completion_length": 49.59375, |
| "epoch": 0.11295681063122924, |
| "grad_norm": 0.3932209610939026, |
| "kl": 0.0052032470703125, |
| "learning_rate": 9.3625e-07, |
| "loss": -0.015652057249099016, |
| "reward": 2.285743832588196, |
| "reward_std": 0.4700127840042114, |
| "rewards/GDino": 0.7444302141666412, |
| "rewards/GIT": 0.5256561636924744, |
| "rewards/HPSv2": 0.2712440490722656, |
| "rewards/ORM": 0.7444134056568146, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.5625, |
| "step": 102 |
| }, |
| { |
| "completion_length": 54.6875, |
| "epoch": 0.11406423034330011, |
| "grad_norm": 0.44802016019821167, |
| "kl": 0.006866455078125, |
| "learning_rate": 9.35625e-07, |
| "loss": 0.004784752381965518, |
| "reward": 1.801784873008728, |
| "reward_std": 0.5139727592468262, |
| "rewards/GDino": 0.6817658245563507, |
| "rewards/GIT": 0.18415232002735138, |
| "rewards/HPSv2": 0.2674713134765625, |
| "rewards/ORM": 0.6683953106403351, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.5, |
| "step": 103 |
| }, |
| { |
| "completion_length": 41.671875, |
| "epoch": 0.11517165005537099, |
| "grad_norm": 0.5610178709030151, |
| "kl": 0.0048675537109375, |
| "learning_rate": 9.35e-07, |
| "loss": 0.012662995606660843, |
| "reward": 2.2165188789367676, |
| "reward_std": 0.4234919399023056, |
| "rewards/GDino": 0.8522021770477295, |
| "rewards/GIT": 0.4973383694887161, |
| "rewards/HPSv2": 0.25607872009277344, |
| "rewards/ORM": 0.6108995676040649, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.875, |
| "step": 104 |
| }, |
| { |
| "completion_length": 48.609375, |
| "epoch": 0.11627906976744186, |
| "grad_norm": 0.5205046534538269, |
| "kl": 0.00438690185546875, |
| "learning_rate": 9.34375e-07, |
| "loss": 0.010773615911602974, |
| "reward": 1.9969267845153809, |
| "reward_std": 0.462581530213356, |
| "rewards/GDino": 0.739062488079071, |
| "rewards/GIT": 0.5269564837217331, |
| "rewards/HPSv2": 0.2528705596923828, |
| "rewards/ORM": 0.4780370891094208, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -20.9375, |
| "step": 105 |
| }, |
| { |
| "completion_length": 53.09375, |
| "epoch": 0.11738648947951273, |
| "grad_norm": 0.4945337772369385, |
| "kl": 0.00521087646484375, |
| "learning_rate": 9.3375e-07, |
| "loss": -0.015708873979747295, |
| "reward": 1.9371621012687683, |
| "reward_std": 0.3034388795495033, |
| "rewards/GDino": 0.7275120615959167, |
| "rewards/GIT": 0.5758572816848755, |
| "rewards/HPSv2": 0.26700592041015625, |
| "rewards/ORM": 0.3667868673801422, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -20.75, |
| "step": 106 |
| }, |
| { |
| "completion_length": 47.890625, |
| "epoch": 0.1184939091915836, |
| "grad_norm": 0.5665311217308044, |
| "kl": 0.007598876953125, |
| "learning_rate": 9.33125e-07, |
| "loss": -0.003110818797722459, |
| "reward": 2.5077059268951416, |
| "reward_std": 0.4203761965036392, |
| "rewards/GDino": 0.9479166865348816, |
| "rewards/GIT": 0.7534970641136169, |
| "rewards/HPSv2": 0.2581634521484375, |
| "rewards/ORM": 0.548128753900528, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -20.6875, |
| "step": 107 |
| }, |
| { |
| "completion_length": 42.859375, |
| "epoch": 0.11960132890365449, |
| "grad_norm": 0.5378500819206238, |
| "kl": 0.0073089599609375, |
| "learning_rate": 9.325e-07, |
| "loss": -0.0038322817999869585, |
| "reward": 2.261056423187256, |
| "reward_std": 0.27708302438259125, |
| "rewards/GDino": 0.7834341526031494, |
| "rewards/GIT": 0.4610650986433029, |
| "rewards/HPSv2": 0.28486061096191406, |
| "rewards/ORM": 0.731696605682373, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -20.75, |
| "step": 108 |
| }, |
| { |
| "completion_length": 50.828125, |
| "epoch": 0.12070874861572536, |
| "grad_norm": 0.4483701288700104, |
| "kl": 0.0067138671875, |
| "learning_rate": 9.31875e-07, |
| "loss": 0.021743599325418472, |
| "reward": 2.118706166744232, |
| "reward_std": 0.3737848997116089, |
| "rewards/GDino": 0.793749988079071, |
| "rewards/GIT": 0.5421168804168701, |
| "rewards/HPSv2": 0.27524757385253906, |
| "rewards/ORM": 0.5075916796922684, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.4375, |
| "step": 109 |
| }, |
| { |
| "completion_length": 44.40625, |
| "epoch": 0.12181616832779624, |
| "grad_norm": 5.2752509117126465, |
| "kl": 0.005615234375, |
| "learning_rate": 9.3125e-07, |
| "loss": -0.008640175685286522, |
| "reward": 2.223414659500122, |
| "reward_std": 0.46220165491104126, |
| "rewards/GDino": 0.8336420953273773, |
| "rewards/GIT": 0.4199307709932327, |
| "rewards/HPSv2": 0.24015045166015625, |
| "rewards/ORM": 0.7296914756298065, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.9375, |
| "step": 110 |
| }, |
| { |
| "completion_length": 54.234375, |
| "epoch": 0.12292358803986711, |
| "grad_norm": 1.4650001525878906, |
| "kl": 0.0050811767578125, |
| "learning_rate": 9.30625e-07, |
| "loss": 0.01813027122989297, |
| "reward": 1.980902910232544, |
| "reward_std": 0.3374909907579422, |
| "rewards/GDino": 0.7292370200157166, |
| "rewards/GIT": 0.3778613805770874, |
| "rewards/HPSv2": 0.27063751220703125, |
| "rewards/ORM": 0.603166937828064, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.875, |
| "step": 111 |
| }, |
| { |
| "completion_length": 42.015625, |
| "epoch": 0.12403100775193798, |
| "grad_norm": 2.8602488040924072, |
| "kl": 0.014862060546875, |
| "learning_rate": 9.3e-07, |
| "loss": 0.001048431033268571, |
| "reward": 2.127329468727112, |
| "reward_std": 0.3023644834756851, |
| "rewards/GDino": 0.7870483100414276, |
| "rewards/GIT": 0.3371267020702362, |
| "rewards/HPSv2": 0.2804222106933594, |
| "rewards/ORM": 0.7227321267127991, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.5625, |
| "step": 112 |
| }, |
| { |
| "completion_length": 40.921875, |
| "epoch": 0.12513842746400886, |
| "grad_norm": 0.47990044951438904, |
| "kl": 0.006256103515625, |
| "learning_rate": 9.293749999999999e-07, |
| "loss": -0.0006602238863706589, |
| "reward": 1.6797617077827454, |
| "reward_std": 0.38670530915260315, |
| "rewards/GDino": 0.6658706367015839, |
| "rewards/GIT": 0.29068493843078613, |
| "rewards/HPSv2": 0.2770404815673828, |
| "rewards/ORM": 0.44616562128067017, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.8125, |
| "step": 113 |
| }, |
| { |
| "completion_length": 45.734375, |
| "epoch": 0.12624584717607973, |
| "grad_norm": 0.7622689008712769, |
| "kl": 0.006622314453125, |
| "learning_rate": 9.287499999999999e-07, |
| "loss": 0.01737637398764491, |
| "reward": 2.336304783821106, |
| "reward_std": 0.2698482424020767, |
| "rewards/GDino": 0.7757812738418579, |
| "rewards/GIT": 0.6553223580121994, |
| "rewards/HPSv2": 0.2794780731201172, |
| "rewards/ORM": 0.6257232129573822, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.6875, |
| "step": 114 |
| }, |
| { |
| "completion_length": 49.65625, |
| "epoch": 0.1273532668881506, |
| "grad_norm": 0.5282474756240845, |
| "kl": 0.007537841796875, |
| "learning_rate": 9.281249999999999e-07, |
| "loss": 0.016482284292578697, |
| "reward": 2.164160192012787, |
| "reward_std": 0.3744830787181854, |
| "rewards/GDino": 0.710269957780838, |
| "rewards/GIT": 0.5369775593280792, |
| "rewards/HPSv2": 0.2497711181640625, |
| "rewards/ORM": 0.6671415567398071, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -20.6875, |
| "step": 115 |
| }, |
| { |
| "completion_length": 51.953125, |
| "epoch": 0.12846068660022147, |
| "grad_norm": 0.477001816034317, |
| "kl": 0.0054931640625, |
| "learning_rate": 9.274999999999999e-07, |
| "loss": 0.005994495470076799, |
| "reward": 2.3729158639907837, |
| "reward_std": 0.29741741716861725, |
| "rewards/GDino": 0.7775339484214783, |
| "rewards/GIT": 0.54022216796875, |
| "rewards/HPSv2": 0.26922607421875, |
| "rewards/ORM": 0.7859334647655487, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -20.5, |
| "step": 116 |
| }, |
| { |
| "completion_length": 47.203125, |
| "epoch": 0.12956810631229235, |
| "grad_norm": 0.5086694955825806, |
| "kl": 0.007476806640625, |
| "learning_rate": 9.268749999999999e-07, |
| "loss": -0.011362070217728615, |
| "reward": 2.5480291843414307, |
| "reward_std": 0.21939973533153534, |
| "rewards/GDino": 0.8421875238418579, |
| "rewards/GIT": 0.6468265354633331, |
| "rewards/HPSv2": 0.28557777404785156, |
| "rewards/ORM": 0.7734375, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.1875, |
| "step": 117 |
| }, |
| { |
| "completion_length": 47.96875, |
| "epoch": 0.13067552602436322, |
| "grad_norm": 0.4937967360019684, |
| "kl": 0.008392333984375, |
| "learning_rate": 9.2625e-07, |
| "loss": 0.021687609143555164, |
| "reward": 1.9581258296966553, |
| "reward_std": 0.4126932621002197, |
| "rewards/GDino": 0.7584865391254425, |
| "rewards/GIT": 0.40825480222702026, |
| "rewards/HPSv2": 0.2484416961669922, |
| "rewards/ORM": 0.5429428219795227, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.25, |
| "step": 118 |
| }, |
| { |
| "completion_length": 42.109375, |
| "epoch": 0.13178294573643412, |
| "grad_norm": 0.5172230005264282, |
| "kl": 0.008087158203125, |
| "learning_rate": 9.25625e-07, |
| "loss": -0.019072898663580418, |
| "reward": 2.281066656112671, |
| "reward_std": 0.5171918570995331, |
| "rewards/GDino": 0.777093768119812, |
| "rewards/GIT": 0.48773056268692017, |
| "rewards/HPSv2": 0.2693214416503906, |
| "rewards/ORM": 0.7469209730625153, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.6875, |
| "step": 119 |
| }, |
| { |
| "completion_length": 47.03125, |
| "epoch": 0.132890365448505, |
| "grad_norm": 0.5932603478431702, |
| "kl": 0.006134033203125, |
| "learning_rate": 9.25e-07, |
| "loss": -0.013699718751013279, |
| "reward": 1.8187614679336548, |
| "reward_std": 0.3037511110305786, |
| "rewards/GDino": 0.6585085391998291, |
| "rewards/GIT": 0.44199828803539276, |
| "rewards/HPSv2": 0.2613792419433594, |
| "rewards/ORM": 0.4568754881620407, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.25, |
| "step": 120 |
| }, |
| { |
| "completion_length": 43.09375, |
| "epoch": 0.13399778516057587, |
| "grad_norm": 0.5679785013198853, |
| "kl": 0.0063629150390625, |
| "learning_rate": 9.243749999999999e-07, |
| "loss": 0.00992331630550325, |
| "reward": 2.001632511615753, |
| "reward_std": 0.43007975816726685, |
| "rewards/GDino": 0.7286458313465118, |
| "rewards/GIT": 0.4185456335544586, |
| "rewards/HPSv2": 0.25841522216796875, |
| "rewards/ORM": 0.5960258543491364, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.4375, |
| "step": 121 |
| }, |
| { |
| "completion_length": 45.125, |
| "epoch": 0.13510520487264674, |
| "grad_norm": 0.5153163075447083, |
| "kl": 0.0077056884765625, |
| "learning_rate": 9.237499999999999e-07, |
| "loss": 0.023711273446679115, |
| "reward": 1.9004549980163574, |
| "reward_std": 0.30516810715198517, |
| "rewards/GDino": 0.7201891243457794, |
| "rewards/GIT": 0.3620809018611908, |
| "rewards/HPSv2": 0.2901439666748047, |
| "rewards/ORM": 0.5280410796403885, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -20.9375, |
| "step": 122 |
| }, |
| { |
| "completion_length": 40.984375, |
| "epoch": 0.1362126245847176, |
| "grad_norm": 0.5419678092002869, |
| "kl": 0.0060882568359375, |
| "learning_rate": 9.23125e-07, |
| "loss": 0.002633487805724144, |
| "reward": 1.8504286408424377, |
| "reward_std": 0.48781776428222656, |
| "rewards/GDino": 0.7166666686534882, |
| "rewards/GIT": 0.484364315867424, |
| "rewards/HPSv2": 0.25711822509765625, |
| "rewards/ORM": 0.39227938652038574, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.375, |
| "step": 123 |
| }, |
| { |
| "completion_length": 54.109375, |
| "epoch": 0.13732004429678848, |
| "grad_norm": 3.3991947174072266, |
| "kl": 0.00714111328125, |
| "learning_rate": 9.225e-07, |
| "loss": 0.010619609151035547, |
| "reward": 2.2070562839508057, |
| "reward_std": 0.4152011424303055, |
| "rewards/GDino": 0.7853051722049713, |
| "rewards/GIT": 0.4775720238685608, |
| "rewards/HPSv2": 0.25039100646972656, |
| "rewards/ORM": 0.6937879621982574, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.625, |
| "step": 124 |
| }, |
| { |
| "completion_length": 52.390625, |
| "epoch": 0.13842746400885936, |
| "grad_norm": 0.6660559773445129, |
| "kl": 0.00579833984375, |
| "learning_rate": 9.21875e-07, |
| "loss": 0.030846341978758574, |
| "reward": 2.0939546823501587, |
| "reward_std": 0.3831700086593628, |
| "rewards/GDino": 0.8140697479248047, |
| "rewards/GIT": 0.5598450750112534, |
| "rewards/HPSv2": 0.24968528747558594, |
| "rewards/ORM": 0.47035445272922516, |
| "self_certainty_semantic": -25.5625, |
| "self_certainty_token": -21.625, |
| "step": 125 |
| }, |
| { |
| "completion_length": 40.546875, |
| "epoch": 0.13953488372093023, |
| "grad_norm": 0.4604582190513611, |
| "kl": 0.0074005126953125, |
| "learning_rate": 9.2125e-07, |
| "loss": -0.00871883099898696, |
| "reward": 1.8355292081832886, |
| "reward_std": 0.3674861043691635, |
| "rewards/GDino": 0.7088627219200134, |
| "rewards/GIT": 0.29635151475667953, |
| "rewards/HPSv2": 0.25839805603027344, |
| "rewards/ORM": 0.5719168931245804, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.125, |
| "step": 126 |
| }, |
| { |
| "completion_length": 54.71875, |
| "epoch": 0.1406423034330011, |
| "grad_norm": 0.8080440759658813, |
| "kl": 0.0090179443359375, |
| "learning_rate": 9.20625e-07, |
| "loss": -0.01499070762656629, |
| "reward": 2.144508123397827, |
| "reward_std": 0.3519645929336548, |
| "rewards/GDino": 0.7293722033500671, |
| "rewards/GIT": 0.42373301088809967, |
| "rewards/HPSv2": 0.27573204040527344, |
| "rewards/ORM": 0.7156709134578705, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.625, |
| "step": 127 |
| }, |
| { |
| "completion_length": 51.46875, |
| "epoch": 0.14174972314507198, |
| "grad_norm": 0.5151174068450928, |
| "kl": 0.007843017578125, |
| "learning_rate": 9.2e-07, |
| "loss": 0.011134594678878784, |
| "reward": 1.9095789790153503, |
| "reward_std": 0.5153420865535736, |
| "rewards/GDino": 0.7141143381595612, |
| "rewards/GIT": 0.26926109194755554, |
| "rewards/HPSv2": 0.26451873779296875, |
| "rewards/ORM": 0.6616848409175873, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.25, |
| "step": 128 |
| }, |
| { |
| "completion_length": 49.1875, |
| "epoch": 0.14285714285714285, |
| "grad_norm": 0.5960005521774292, |
| "kl": 0.0080413818359375, |
| "learning_rate": 9.19375e-07, |
| "loss": -0.0018172780983150005, |
| "reward": 2.0566558837890625, |
| "reward_std": 0.29967472702264786, |
| "rewards/GDino": 0.7159374058246613, |
| "rewards/GIT": 0.4233300983905792, |
| "rewards/HPSv2": 0.25041770935058594, |
| "rewards/ORM": 0.6669707000255585, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.0625, |
| "step": 129 |
| }, |
| { |
| "completion_length": 47.046875, |
| "epoch": 0.14396456256921372, |
| "grad_norm": 0.5405691862106323, |
| "kl": 0.0067596435546875, |
| "learning_rate": 9.187499999999999e-07, |
| "loss": 0.032197900116443634, |
| "reward": 1.9141977429389954, |
| "reward_std": 0.455816388130188, |
| "rewards/GDino": 0.7374999821186066, |
| "rewards/GIT": 0.4731251299381256, |
| "rewards/HPSv2": 0.2379016876220703, |
| "rewards/ORM": 0.46567094326019287, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.1875, |
| "step": 130 |
| }, |
| { |
| "completion_length": 44.125, |
| "epoch": 0.1450719822812846, |
| "grad_norm": 0.524912416934967, |
| "kl": 0.009796142578125, |
| "learning_rate": 9.181249999999999e-07, |
| "loss": -0.0008213929831981659, |
| "reward": 1.8893061876296997, |
| "reward_std": 0.3467349708080292, |
| "rewards/GDino": 0.7044448256492615, |
| "rewards/GIT": 0.35102422535419464, |
| "rewards/HPSv2": 0.26580047607421875, |
| "rewards/ORM": 0.5680365860462189, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.8125, |
| "step": 131 |
| }, |
| { |
| "completion_length": 34.625, |
| "epoch": 0.1461794019933555, |
| "grad_norm": 0.8808599710464478, |
| "kl": 0.011566162109375, |
| "learning_rate": 9.174999999999999e-07, |
| "loss": -0.003930883482098579, |
| "reward": 2.177262306213379, |
| "reward_std": 0.5171742737293243, |
| "rewards/GDino": 0.8225250542163849, |
| "rewards/GIT": 0.4376496821641922, |
| "rewards/HPSv2": 0.28387451171875, |
| "rewards/ORM": 0.6332131624221802, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.8125, |
| "step": 132 |
| }, |
| { |
| "completion_length": 48.515625, |
| "epoch": 0.14728682170542637, |
| "grad_norm": 0.5650766491889954, |
| "kl": 0.006622314453125, |
| "learning_rate": 9.168749999999999e-07, |
| "loss": -0.00018032779917120934, |
| "reward": 2.5767033100128174, |
| "reward_std": 0.21334625780582428, |
| "rewards/GDino": 0.768750011920929, |
| "rewards/GIT": 0.6930812895298004, |
| "rewards/HPSv2": 0.2711219787597656, |
| "rewards/ORM": 0.8437499701976776, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -21.3125, |
| "step": 133 |
| }, |
| { |
| "completion_length": 44.640625, |
| "epoch": 0.14839424141749724, |
| "grad_norm": 0.5156822800636292, |
| "kl": 0.004974365234375, |
| "learning_rate": 9.1625e-07, |
| "loss": 0.002806268632411957, |
| "reward": 1.9863407611846924, |
| "reward_std": 0.49281907081604004, |
| "rewards/GDino": 0.7205729484558105, |
| "rewards/GIT": 0.500615194439888, |
| "rewards/HPSv2": 0.25568580627441406, |
| "rewards/ORM": 0.5094669014215469, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -20.5, |
| "step": 134 |
| }, |
| { |
| "completion_length": 45.390625, |
| "epoch": 0.14950166112956811, |
| "grad_norm": 0.8464062809944153, |
| "kl": 0.0070343017578125, |
| "learning_rate": 9.15625e-07, |
| "loss": -0.019531114026904106, |
| "reward": 2.0325437784194946, |
| "reward_std": 0.40662893652915955, |
| "rewards/GDino": 0.7845472693443298, |
| "rewards/GIT": 0.48149144649505615, |
| "rewards/HPSv2": 0.2637767791748047, |
| "rewards/ORM": 0.5027283281087875, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -20.3125, |
| "step": 135 |
| }, |
| { |
| "completion_length": 51.34375, |
| "epoch": 0.150609080841639, |
| "grad_norm": 0.5737318992614746, |
| "kl": 0.0070343017578125, |
| "learning_rate": 9.15e-07, |
| "loss": -0.0035284715704619884, |
| "reward": 1.6841511130332947, |
| "reward_std": 0.3555753082036972, |
| "rewards/GDino": 0.6274834871292114, |
| "rewards/GIT": 0.27718986570835114, |
| "rewards/HPSv2": 0.2579154968261719, |
| "rewards/ORM": 0.521562248468399, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.0, |
| "step": 136 |
| }, |
| { |
| "completion_length": 44.890625, |
| "epoch": 0.15171650055370986, |
| "grad_norm": 0.9722269773483276, |
| "kl": 0.009033203125, |
| "learning_rate": 9.14375e-07, |
| "loss": -0.010386745911091566, |
| "reward": 2.115400731563568, |
| "reward_std": 0.4073975533246994, |
| "rewards/GDino": 0.7504827678203583, |
| "rewards/GIT": 0.5390563532710075, |
| "rewards/HPSv2": 0.25206947326660156, |
| "rewards/ORM": 0.5737921595573425, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.625, |
| "step": 137 |
| }, |
| { |
| "completion_length": 45.75, |
| "epoch": 0.15282392026578073, |
| "grad_norm": 0.7052740454673767, |
| "kl": 0.010101318359375, |
| "learning_rate": 9.137499999999999e-07, |
| "loss": -0.009516147896647453, |
| "reward": 1.8091301918029785, |
| "reward_std": 0.3067747950553894, |
| "rewards/GDino": 0.6395186185836792, |
| "rewards/GIT": 0.19299907237291336, |
| "rewards/HPSv2": 0.26455116271972656, |
| "rewards/ORM": 0.7120613753795624, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.1875, |
| "step": 138 |
| }, |
| { |
| "completion_length": 52.609375, |
| "epoch": 0.1539313399778516, |
| "grad_norm": 0.49324488639831543, |
| "kl": 0.0060577392578125, |
| "learning_rate": 9.131249999999999e-07, |
| "loss": 0.019678042270243168, |
| "reward": 2.08823698759079, |
| "reward_std": 0.3092179298400879, |
| "rewards/GDino": 0.7869435846805573, |
| "rewards/GIT": 0.3419180363416672, |
| "rewards/HPSv2": 0.27410125732421875, |
| "rewards/ORM": 0.6852740943431854, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.5625, |
| "step": 139 |
| }, |
| { |
| "completion_length": 44.65625, |
| "epoch": 0.15503875968992248, |
| "grad_norm": 0.46719232201576233, |
| "kl": 0.005218505859375, |
| "learning_rate": 9.124999999999999e-07, |
| "loss": -0.016016804613173008, |
| "reward": 2.2374342679977417, |
| "reward_std": 0.4363926351070404, |
| "rewards/GDino": 0.8431436419487, |
| "rewards/GIT": 0.5656594336032867, |
| "rewards/HPSv2": 0.25456809997558594, |
| "rewards/ORM": 0.5740630030632019, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.5, |
| "step": 140 |
| }, |
| { |
| "completion_length": 58.671875, |
| "epoch": 0.15614617940199335, |
| "grad_norm": 0.4546290934085846, |
| "kl": 0.004730224609375, |
| "learning_rate": 9.11875e-07, |
| "loss": -0.0007884092628955841, |
| "reward": 2.369232416152954, |
| "reward_std": 0.263886496424675, |
| "rewards/GDino": 0.7109375298023224, |
| "rewards/GIT": 0.7504715323448181, |
| "rewards/HPSv2": 0.245635986328125, |
| "rewards/ORM": 0.6621872782707214, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -20.75, |
| "step": 141 |
| }, |
| { |
| "completion_length": 42.4375, |
| "epoch": 0.15725359911406422, |
| "grad_norm": 0.9214792251586914, |
| "kl": 0.009521484375, |
| "learning_rate": 9.1125e-07, |
| "loss": 0.003012734232470393, |
| "reward": 2.1930705904960632, |
| "reward_std": 0.39259086549282074, |
| "rewards/GDino": 0.7755208611488342, |
| "rewards/GIT": 0.5075857639312744, |
| "rewards/HPSv2": 0.2599220275878906, |
| "rewards/ORM": 0.6500419527292252, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -20.625, |
| "step": 142 |
| }, |
| { |
| "completion_length": 44.984375, |
| "epoch": 0.1583610188261351, |
| "grad_norm": 0.622131884098053, |
| "kl": 0.007476806640625, |
| "learning_rate": 9.10625e-07, |
| "loss": 0.012470124522224069, |
| "reward": 1.945671796798706, |
| "reward_std": 0.3904338628053665, |
| "rewards/GDino": 0.7124259173870087, |
| "rewards/GIT": 0.36576879024505615, |
| "rewards/HPSv2": 0.2677898406982422, |
| "rewards/ORM": 0.599687248468399, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -20.5625, |
| "step": 143 |
| }, |
| { |
| "completion_length": 50.4375, |
| "epoch": 0.15946843853820597, |
| "grad_norm": 0.5398712158203125, |
| "kl": 0.007080078125, |
| "learning_rate": 9.1e-07, |
| "loss": 0.009707295335829258, |
| "reward": 1.9473342895507812, |
| "reward_std": 0.411212295293808, |
| "rewards/GDino": 0.6565104126930237, |
| "rewards/GIT": 0.4918062835931778, |
| "rewards/HPSv2": 0.24593448638916016, |
| "rewards/ORM": 0.5530830472707748, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.1875, |
| "step": 144 |
| }, |
| { |
| "completion_length": 48.828125, |
| "epoch": 0.16057585825027684, |
| "grad_norm": 0.44240400195121765, |
| "kl": 0.0082550048828125, |
| "learning_rate": 9.09375e-07, |
| "loss": -0.015047748805955052, |
| "reward": 1.8875170946121216, |
| "reward_std": 0.37340451776981354, |
| "rewards/GDino": 0.6789085865020752, |
| "rewards/GIT": 0.31243064999580383, |
| "rewards/HPSv2": 0.2805948257446289, |
| "rewards/ORM": 0.6155830323696136, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.375, |
| "step": 145 |
| }, |
| { |
| "completion_length": 42.234375, |
| "epoch": 0.16168327796234774, |
| "grad_norm": 0.6678792238235474, |
| "kl": 0.0080718994140625, |
| "learning_rate": 9.087499999999999e-07, |
| "loss": 0.015477177686989307, |
| "reward": 2.123531699180603, |
| "reward_std": 0.4475431591272354, |
| "rewards/GDino": 0.8569894731044769, |
| "rewards/GIT": 0.35753606259822845, |
| "rewards/HPSv2": 0.26838111877441406, |
| "rewards/ORM": 0.640625, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.5, |
| "step": 146 |
| }, |
| { |
| "completion_length": 47.921875, |
| "epoch": 0.16279069767441862, |
| "grad_norm": 0.4761756658554077, |
| "kl": 0.007293701171875, |
| "learning_rate": 9.081249999999999e-07, |
| "loss": -0.008836451917886734, |
| "reward": 1.6021055579185486, |
| "reward_std": 0.30197490751743317, |
| "rewards/GDino": 0.6556249558925629, |
| "rewards/GIT": 0.1688888967037201, |
| "rewards/HPSv2": 0.24634170532226562, |
| "rewards/ORM": 0.5312500298023224, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.0, |
| "step": 147 |
| }, |
| { |
| "completion_length": 46.921875, |
| "epoch": 0.1638981173864895, |
| "grad_norm": 0.6984077095985413, |
| "kl": 0.011322021484375, |
| "learning_rate": 9.074999999999999e-07, |
| "loss": -0.001696310006082058, |
| "reward": 2.1800928115844727, |
| "reward_std": 0.22076455503702164, |
| "rewards/GDino": 0.8397657871246338, |
| "rewards/GIT": 0.3664309233427048, |
| "rewards/HPSv2": 0.2707710266113281, |
| "rewards/ORM": 0.703125, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.625, |
| "step": 148 |
| }, |
| { |
| "completion_length": 42.984375, |
| "epoch": 0.16500553709856036, |
| "grad_norm": 0.6648632884025574, |
| "kl": 0.008514404296875, |
| "learning_rate": 9.068749999999999e-07, |
| "loss": 0.0001570945605635643, |
| "reward": 2.0893077850341797, |
| "reward_std": 0.40299197286367416, |
| "rewards/GDino": 0.7465624809265137, |
| "rewards/GIT": 0.32270002365112305, |
| "rewards/HPSv2": 0.2650909423828125, |
| "rewards/ORM": 0.7549542784690857, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.0, |
| "step": 149 |
| }, |
| { |
| "completion_length": 46.765625, |
| "epoch": 0.16611295681063123, |
| "grad_norm": 0.4624801278114319, |
| "kl": 0.00653076171875, |
| "learning_rate": 9.0625e-07, |
| "loss": -0.00981504051014781, |
| "reward": 1.8526134490966797, |
| "reward_std": 0.49954167008399963, |
| "rewards/GDino": 0.7183263897895813, |
| "rewards/GIT": 0.28864332288503647, |
| "rewards/HPSv2": 0.2530975341796875, |
| "rewards/ORM": 0.5925461798906326, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.3125, |
| "step": 150 |
| }, |
| { |
| "completion_length": 44.265625, |
| "epoch": 0.1672203765227021, |
| "grad_norm": 0.6167420744895935, |
| "kl": 0.008636474609375, |
| "learning_rate": 9.05625e-07, |
| "loss": 0.015115905553102493, |
| "reward": 2.337567687034607, |
| "reward_std": 0.42467789351940155, |
| "rewards/GDino": 0.8551518619060516, |
| "rewards/GIT": 0.40152132511138916, |
| "rewards/HPSv2": 0.2702198028564453, |
| "rewards/ORM": 0.8106746673583984, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.25, |
| "step": 151 |
| }, |
| { |
| "completion_length": 49.578125, |
| "epoch": 0.16832779623477298, |
| "grad_norm": 0.8095238208770752, |
| "kl": 0.006683349609375, |
| "learning_rate": 9.05e-07, |
| "loss": -0.005870660301297903, |
| "reward": 2.1157608032226562, |
| "reward_std": 0.20268037915229797, |
| "rewards/GDino": 0.8022373914718628, |
| "rewards/GIT": 0.4189887195825577, |
| "rewards/HPSv2": 0.27895164489746094, |
| "rewards/ORM": 0.6155830323696136, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.625, |
| "step": 152 |
| }, |
| { |
| "completion_length": 41.84375, |
| "epoch": 0.16943521594684385, |
| "grad_norm": 0.5198939442634583, |
| "kl": 0.0080718994140625, |
| "learning_rate": 9.04375e-07, |
| "loss": -0.014349173055961728, |
| "reward": 2.219430923461914, |
| "reward_std": 0.31192296743392944, |
| "rewards/GDino": 0.7789298295974731, |
| "rewards/GIT": 0.6309380829334259, |
| "rewards/HPSv2": 0.26394176483154297, |
| "rewards/ORM": 0.5456212162971497, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.125, |
| "step": 153 |
| }, |
| { |
| "completion_length": 39.796875, |
| "epoch": 0.17054263565891473, |
| "grad_norm": 0.4530474543571472, |
| "kl": 0.00823974609375, |
| "learning_rate": 9.0375e-07, |
| "loss": -0.009022563113830984, |
| "reward": 1.62737375497818, |
| "reward_std": 0.3316381424665451, |
| "rewards/GDino": 0.753333568572998, |
| "rewards/GIT": 0.17652657628059387, |
| "rewards/HPSv2": 0.2600135803222656, |
| "rewards/ORM": 0.4375, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.25, |
| "step": 154 |
| }, |
| { |
| "completion_length": 56.828125, |
| "epoch": 0.1716500553709856, |
| "grad_norm": 0.5859449505805969, |
| "kl": 0.01031494140625, |
| "learning_rate": 9.031249999999999e-07, |
| "loss": -0.006706917891278863, |
| "reward": 2.134092330932617, |
| "reward_std": 0.3308701366186142, |
| "rewards/GDino": 0.6453125327825546, |
| "rewards/GIT": 0.44731535762548447, |
| "rewards/HPSv2": 0.26494789123535156, |
| "rewards/ORM": 0.7765165567398071, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -20.6875, |
| "step": 155 |
| }, |
| { |
| "completion_length": 41.078125, |
| "epoch": 0.17275747508305647, |
| "grad_norm": 0.5353882908821106, |
| "kl": 0.010894775390625, |
| "learning_rate": 9.024999999999999e-07, |
| "loss": 0.0037739332765340805, |
| "reward": 1.8532127141952515, |
| "reward_std": 0.3209614157676697, |
| "rewards/GDino": 0.6619158685207367, |
| "rewards/GIT": 0.23281337320804596, |
| "rewards/HPSv2": 0.2831287384033203, |
| "rewards/ORM": 0.6753546595573425, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.125, |
| "step": 156 |
| }, |
| { |
| "completion_length": 54.296875, |
| "epoch": 0.17386489479512734, |
| "grad_norm": 0.45864060521125793, |
| "kl": 0.0081787109375, |
| "learning_rate": 9.018749999999999e-07, |
| "loss": 0.01829966064542532, |
| "reward": 2.4461565017700195, |
| "reward_std": 0.3625805824995041, |
| "rewards/GDino": 0.8224999904632568, |
| "rewards/GIT": 0.7057079672813416, |
| "rewards/HPSv2": 0.24219322204589844, |
| "rewards/ORM": 0.6757553368806839, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.8125, |
| "step": 157 |
| }, |
| { |
| "completion_length": 35.96875, |
| "epoch": 0.17497231450719822, |
| "grad_norm": 0.6962554454803467, |
| "kl": 0.01031494140625, |
| "learning_rate": 9.0125e-07, |
| "loss": 0.018664106726646423, |
| "reward": 2.2535247802734375, |
| "reward_std": 0.3470146059989929, |
| "rewards/GDino": 0.8081650137901306, |
| "rewards/GIT": 0.43000543117523193, |
| "rewards/HPSv2": 0.2705402374267578, |
| "rewards/ORM": 0.7448140382766724, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.375, |
| "step": 158 |
| }, |
| { |
| "completion_length": 48.421875, |
| "epoch": 0.1760797342192691, |
| "grad_norm": 0.7126027941703796, |
| "kl": 0.0084686279296875, |
| "learning_rate": 9.00625e-07, |
| "loss": -0.01690885704010725, |
| "reward": 2.103760838508606, |
| "reward_std": 0.42456358671188354, |
| "rewards/GDino": 0.7740625143051147, |
| "rewards/GIT": 0.5006528943777084, |
| "rewards/HPSv2": 0.2603874206542969, |
| "rewards/ORM": 0.5686581134796143, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -20.8125, |
| "step": 159 |
| }, |
| { |
| "completion_length": 47.15625, |
| "epoch": 0.17718715393134, |
| "grad_norm": 0.4899819791316986, |
| "kl": 0.011016845703125, |
| "learning_rate": 9e-07, |
| "loss": 0.011189845390617847, |
| "reward": 1.9280957579612732, |
| "reward_std": 0.3808829113841057, |
| "rewards/GDino": 0.7389523684978485, |
| "rewards/GIT": 0.29086190462112427, |
| "rewards/HPSv2": 0.27908897399902344, |
| "rewards/ORM": 0.6191926002502441, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.875, |
| "step": 160 |
| }, |
| { |
| "completion_length": 48.0, |
| "epoch": 0.17829457364341086, |
| "grad_norm": 0.8614944815635681, |
| "kl": 0.010162353515625, |
| "learning_rate": 8.99375e-07, |
| "loss": -0.0047345394268631935, |
| "reward": 2.126296818256378, |
| "reward_std": 0.4349767565727234, |
| "rewards/GDino": 0.6996158957481384, |
| "rewards/GIT": 0.4218016564846039, |
| "rewards/HPSv2": 0.26108741760253906, |
| "rewards/ORM": 0.743791937828064, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -20.75, |
| "step": 161 |
| }, |
| { |
| "completion_length": 57.40625, |
| "epoch": 0.17940199335548174, |
| "grad_norm": 0.4764331877231598, |
| "kl": 0.00958251953125, |
| "learning_rate": 8.9875e-07, |
| "loss": -0.02189162978902459, |
| "reward": 2.1861079335212708, |
| "reward_std": 0.2511523813009262, |
| "rewards/GDino": 0.9083333313465118, |
| "rewards/GIT": 0.5160972326993942, |
| "rewards/HPSv2": 0.2581977844238281, |
| "rewards/ORM": 0.5034796893596649, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.5, |
| "step": 162 |
| }, |
| { |
| "completion_length": 41.796875, |
| "epoch": 0.1805094130675526, |
| "grad_norm": 0.63126540184021, |
| "kl": 0.01348876953125, |
| "learning_rate": 8.981249999999999e-07, |
| "loss": -0.010512399720028043, |
| "reward": 2.1882660388946533, |
| "reward_std": 0.26342111825942993, |
| "rewards/GDino": 0.7701247036457062, |
| "rewards/GIT": 0.637158066034317, |
| "rewards/HPSv2": 0.24700498580932617, |
| "rewards/ORM": 0.5339783430099487, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -20.625, |
| "step": 163 |
| }, |
| { |
| "completion_length": 43.78125, |
| "epoch": 0.18161683277962348, |
| "grad_norm": 0.4358772039413452, |
| "kl": 0.01190185546875, |
| "learning_rate": 8.974999999999999e-07, |
| "loss": -0.006003182148560882, |
| "reward": 2.6546283960342407, |
| "reward_std": 0.3610512763261795, |
| "rewards/GDino": 0.8648440539836884, |
| "rewards/GIT": 0.7706953585147858, |
| "rewards/HPSv2": 0.269439697265625, |
| "rewards/ORM": 0.749649316072464, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -20.6875, |
| "step": 164 |
| }, |
| { |
| "completion_length": 48.09375, |
| "epoch": 0.18272425249169436, |
| "grad_norm": 1.585694432258606, |
| "kl": 0.01043701171875, |
| "learning_rate": 8.96875e-07, |
| "loss": 8.579343557357788e-05, |
| "reward": 2.331842541694641, |
| "reward_std": 0.3782896548509598, |
| "rewards/GDino": 0.7984375357627869, |
| "rewards/GIT": 0.5101843625307083, |
| "rewards/HPSv2": 0.27196693420410156, |
| "rewards/ORM": 0.751253753900528, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.8125, |
| "step": 165 |
| }, |
| { |
| "completion_length": 45.34375, |
| "epoch": 0.18383167220376523, |
| "grad_norm": 0.45005306601524353, |
| "kl": 0.01312255859375, |
| "learning_rate": 8.9625e-07, |
| "loss": -0.006219237111508846, |
| "reward": 2.192749261856079, |
| "reward_std": 0.2947642654180527, |
| "rewards/GDino": 0.7351250052452087, |
| "rewards/GIT": 0.45546063780784607, |
| "rewards/HPSv2": 0.24868392944335938, |
| "rewards/ORM": 0.7534796893596649, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.25, |
| "step": 166 |
| }, |
| { |
| "completion_length": 43.015625, |
| "epoch": 0.1849390919158361, |
| "grad_norm": 0.5231027007102966, |
| "kl": 0.010772705078125, |
| "learning_rate": 8.95625e-07, |
| "loss": -0.007611713605001569, |
| "reward": 2.0297417044639587, |
| "reward_std": 0.4215656816959381, |
| "rewards/GDino": 0.7420200109481812, |
| "rewards/GIT": 0.5049543976783752, |
| "rewards/HPSv2": 0.26178741455078125, |
| "rewards/ORM": 0.5209799110889435, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.75, |
| "step": 167 |
| }, |
| { |
| "completion_length": 53.46875, |
| "epoch": 0.18604651162790697, |
| "grad_norm": 0.4332805871963501, |
| "kl": 0.0062408447265625, |
| "learning_rate": 8.95e-07, |
| "loss": -0.004800099181011319, |
| "reward": 2.05144202709198, |
| "reward_std": 0.3129463642835617, |
| "rewards/GDino": 0.7780522406101227, |
| "rewards/GIT": 0.5233069062232971, |
| "rewards/HPSv2": 0.2612953186035156, |
| "rewards/ORM": 0.4887876957654953, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.1875, |
| "step": 168 |
| }, |
| { |
| "completion_length": 37.90625, |
| "epoch": 0.18715393133997785, |
| "grad_norm": 0.45867836475372314, |
| "kl": 0.01751708984375, |
| "learning_rate": 8.94375e-07, |
| "loss": 0.004684945801272988, |
| "reward": 2.2036190032958984, |
| "reward_std": 0.26221713423728943, |
| "rewards/GDino": 0.8604569435119629, |
| "rewards/GIT": 0.4576933681964874, |
| "rewards/HPSv2": 0.28287315368652344, |
| "rewards/ORM": 0.6025954186916351, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.5, |
| "step": 169 |
| }, |
| { |
| "completion_length": 41.375, |
| "epoch": 0.18826135105204872, |
| "grad_norm": 0.4206317365169525, |
| "kl": 0.009979248046875, |
| "learning_rate": 8.9375e-07, |
| "loss": 0.0024775206111371517, |
| "reward": 1.943125069141388, |
| "reward_std": 0.4646635055541992, |
| "rewards/GDino": 0.7305906116962433, |
| "rewards/GIT": 0.4165241867303848, |
| "rewards/HPSv2": 0.24480247497558594, |
| "rewards/ORM": 0.5512078106403351, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.125, |
| "step": 170 |
| }, |
| { |
| "completion_length": 40.703125, |
| "epoch": 0.1893687707641196, |
| "grad_norm": 0.697102963924408, |
| "kl": 0.010833740234375, |
| "learning_rate": 8.931249999999999e-07, |
| "loss": -0.003134746104478836, |
| "reward": 2.203751564025879, |
| "reward_std": 0.3228776603937149, |
| "rewards/GDino": 0.7603735029697418, |
| "rewards/GIT": 0.5309787690639496, |
| "rewards/HPSv2": 0.2583751678466797, |
| "rewards/ORM": 0.654024064540863, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.1875, |
| "step": 171 |
| }, |
| { |
| "completion_length": 50.515625, |
| "epoch": 0.19047619047619047, |
| "grad_norm": 0.5617944002151489, |
| "kl": 0.008392333984375, |
| "learning_rate": 8.924999999999999e-07, |
| "loss": 0.03539674496278167, |
| "reward": 1.7965713739395142, |
| "reward_std": 0.3860231041908264, |
| "rewards/GDino": 0.6837728917598724, |
| "rewards/GIT": 0.4299345314502716, |
| "rewards/HPSv2": 0.2496967315673828, |
| "rewards/ORM": 0.4331671893596649, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.25, |
| "step": 172 |
| }, |
| { |
| "completion_length": 35.5625, |
| "epoch": 0.19158361018826134, |
| "grad_norm": 0.4585064649581909, |
| "kl": 0.017242431640625, |
| "learning_rate": 8.918749999999999e-07, |
| "loss": 0.002358448226004839, |
| "reward": 2.1448813676834106, |
| "reward_std": 0.31261830031871796, |
| "rewards/GDino": 0.834684431552887, |
| "rewards/GIT": 0.36106863617897034, |
| "rewards/HPSv2": 0.28739356994628906, |
| "rewards/ORM": 0.6617347896099091, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.5625, |
| "step": 173 |
| }, |
| { |
| "completion_length": 46.859375, |
| "epoch": 0.19269102990033224, |
| "grad_norm": 0.42562612891197205, |
| "kl": 0.00872802734375, |
| "learning_rate": 8.912499999999999e-07, |
| "loss": 0.0036601885221898556, |
| "reward": 2.2070860862731934, |
| "reward_std": 0.3847656697034836, |
| "rewards/GDino": 0.7723565697669983, |
| "rewards/GIT": 0.5430227518081665, |
| "rewards/HPSv2": 0.25701904296875, |
| "rewards/ORM": 0.634687751531601, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -20.875, |
| "step": 174 |
| }, |
| { |
| "completion_length": 46.0, |
| "epoch": 0.1937984496124031, |
| "grad_norm": 0.5808805227279663, |
| "kl": 0.0078277587890625, |
| "learning_rate": 8.906249999999999e-07, |
| "loss": 0.032253723591566086, |
| "reward": 2.157870888710022, |
| "reward_std": 0.2512262612581253, |
| "rewards/GDino": 0.7799479365348816, |
| "rewards/GIT": 0.5305047780275345, |
| "rewards/HPSv2": 0.2598762512207031, |
| "rewards/ORM": 0.587541937828064, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.125, |
| "step": 175 |
| }, |
| { |
| "completion_length": 58.375, |
| "epoch": 0.19490586932447398, |
| "grad_norm": 0.5789065957069397, |
| "kl": 0.0071868896484375, |
| "learning_rate": 8.9e-07, |
| "loss": -0.014507739804685116, |
| "reward": 1.955940306186676, |
| "reward_std": 0.3165567219257355, |
| "rewards/GDino": 0.7864583432674408, |
| "rewards/GIT": 0.4266613572835922, |
| "rewards/HPSv2": 0.25844573974609375, |
| "rewards/ORM": 0.484375, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -20.9375, |
| "step": 176 |
| }, |
| { |
| "completion_length": 51.328125, |
| "epoch": 0.19601328903654486, |
| "grad_norm": 1.309583067893982, |
| "kl": 0.013214111328125, |
| "learning_rate": 8.89375e-07, |
| "loss": -0.0020376548636704683, |
| "reward": 2.2406471967697144, |
| "reward_std": 0.3685739040374756, |
| "rewards/GDino": 0.7480616569519043, |
| "rewards/GIT": 0.4961736798286438, |
| "rewards/HPSv2": 0.25730323791503906, |
| "rewards/ORM": 0.7391084432601929, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -22.25, |
| "step": 177 |
| }, |
| { |
| "completion_length": 56.609375, |
| "epoch": 0.19712070874861573, |
| "grad_norm": 0.6938892602920532, |
| "kl": 0.008270263671875, |
| "learning_rate": 8.8875e-07, |
| "loss": 0.005612233653664589, |
| "reward": 2.1296470165252686, |
| "reward_std": 0.3742387443780899, |
| "rewards/GDino": 0.7609374821186066, |
| "rewards/GIT": 0.4125446677207947, |
| "rewards/HPSv2": 0.2701892852783203, |
| "rewards/ORM": 0.6859754621982574, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.125, |
| "step": 178 |
| }, |
| { |
| "completion_length": 37.234375, |
| "epoch": 0.1982281284606866, |
| "grad_norm": 0.4790467917919159, |
| "kl": 0.006927490234375, |
| "learning_rate": 8.88125e-07, |
| "loss": 0.01884503196924925, |
| "reward": 2.431138515472412, |
| "reward_std": 0.24652785062789917, |
| "rewards/GDino": 0.8910974562168121, |
| "rewards/GIT": 0.5796742737293243, |
| "rewards/HPSv2": 0.28255462646484375, |
| "rewards/ORM": 0.677812248468399, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.375, |
| "step": 179 |
| }, |
| { |
| "completion_length": 45.921875, |
| "epoch": 0.19933554817275748, |
| "grad_norm": 0.4324551522731781, |
| "kl": 0.011749267578125, |
| "learning_rate": 8.874999999999999e-07, |
| "loss": -0.016566987615078688, |
| "reward": 1.7641431093215942, |
| "reward_std": 0.42914003133773804, |
| "rewards/GDino": 0.7295474410057068, |
| "rewards/GIT": 0.32154107093811035, |
| "rewards/HPSv2": 0.25917816162109375, |
| "rewards/ORM": 0.4538763463497162, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.5, |
| "step": 180 |
| }, |
| { |
| "completion_length": 44.921875, |
| "epoch": 0.20044296788482835, |
| "grad_norm": 0.7271856665611267, |
| "kl": 0.008270263671875, |
| "learning_rate": 8.86875e-07, |
| "loss": -0.01033696997910738, |
| "reward": 1.8466984629631042, |
| "reward_std": 0.33254362642765045, |
| "rewards/GDino": 0.6390625238418579, |
| "rewards/GIT": 0.35668525099754333, |
| "rewards/HPSv2": 0.27863311767578125, |
| "rewards/ORM": 0.5723176002502441, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.125, |
| "step": 181 |
| }, |
| { |
| "completion_length": 40.84375, |
| "epoch": 0.20155038759689922, |
| "grad_norm": 0.5530015230178833, |
| "kl": 0.02142333984375, |
| "learning_rate": 8.8625e-07, |
| "loss": -0.008886129595339298, |
| "reward": 1.7842278480529785, |
| "reward_std": 0.4365523010492325, |
| "rewards/GDino": 0.7104989886283875, |
| "rewards/GIT": 0.34373709559440613, |
| "rewards/HPSv2": 0.27834129333496094, |
| "rewards/ORM": 0.4516504108905792, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0, |
| "step": 182 |
| }, |
| { |
| "completion_length": 51.34375, |
| "epoch": 0.2026578073089701, |
| "grad_norm": 0.545352041721344, |
| "kl": 0.015411376953125, |
| "learning_rate": 8.85625e-07, |
| "loss": -0.010915862862020731, |
| "reward": 1.9055233001708984, |
| "reward_std": 0.38334617018699646, |
| "rewards/GDino": 0.7308869063854218, |
| "rewards/GIT": 0.3478083163499832, |
| "rewards/HPSv2": 0.2643280029296875, |
| "rewards/ORM": 0.5625000298023224, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.5, |
| "step": 183 |
| }, |
| { |
| "completion_length": 38.890625, |
| "epoch": 0.20376522702104097, |
| "grad_norm": 0.5411937832832336, |
| "kl": 0.01177978515625, |
| "learning_rate": 8.85e-07, |
| "loss": 0.013757664943113923, |
| "reward": 2.716238021850586, |
| "reward_std": 0.29777073860168457, |
| "rewards/GDino": 0.925000011920929, |
| "rewards/GIT": 0.7263616919517517, |
| "rewards/HPSv2": 0.2595672607421875, |
| "rewards/ORM": 0.805308997631073, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.4375, |
| "step": 184 |
| }, |
| { |
| "completion_length": 41.65625, |
| "epoch": 0.20487264673311184, |
| "grad_norm": 1.4033399820327759, |
| "kl": 0.016143798828125, |
| "learning_rate": 8.84375e-07, |
| "loss": 0.016424793750047684, |
| "reward": 2.2067692279815674, |
| "reward_std": 0.44875267148017883, |
| "rewards/GDino": 0.8457056879997253, |
| "rewards/GIT": 0.31159064173698425, |
| "rewards/HPSv2": 0.275634765625, |
| "rewards/ORM": 0.7738381326198578, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.375, |
| "step": 185 |
| }, |
| { |
| "completion_length": 40.21875, |
| "epoch": 0.2059800664451827, |
| "grad_norm": 0.545001208782196, |
| "kl": 0.0111083984375, |
| "learning_rate": 8.8375e-07, |
| "loss": 0.010479988530278206, |
| "reward": 1.9079334735870361, |
| "reward_std": 0.3735552281141281, |
| "rewards/GDino": 0.6260845363140106, |
| "rewards/GIT": 0.33919139206409454, |
| "rewards/HPSv2": 0.2270364761352539, |
| "rewards/ORM": 0.7156209945678711, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.375, |
| "step": 186 |
| }, |
| { |
| "completion_length": 42.25, |
| "epoch": 0.2070874861572536, |
| "grad_norm": 0.5201876759529114, |
| "kl": 0.012786865234375, |
| "learning_rate": 8.83125e-07, |
| "loss": -0.005474693141877651, |
| "reward": 2.160323202610016, |
| "reward_std": 0.380416139960289, |
| "rewards/GDino": 0.8049721121788025, |
| "rewards/GIT": 0.5423709452152252, |
| "rewards/HPSv2": 0.25914573669433594, |
| "rewards/ORM": 0.5538343787193298, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -20.625, |
| "step": 187 |
| }, |
| { |
| "completion_length": 41.546875, |
| "epoch": 0.2081949058693245, |
| "grad_norm": 1.0598704814910889, |
| "kl": 0.0123291015625, |
| "learning_rate": 8.824999999999999e-07, |
| "loss": -0.005072480300441384, |
| "reward": 2.198424220085144, |
| "reward_std": 0.318149596452713, |
| "rewards/GDino": 0.77506023645401, |
| "rewards/GIT": 0.2563297525048256, |
| "rewards/HPSv2": 0.26827430725097656, |
| "rewards/ORM": 0.8987601101398468, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.6875, |
| "step": 188 |
| }, |
| { |
| "completion_length": 45.859375, |
| "epoch": 0.20930232558139536, |
| "grad_norm": 1.0103754997253418, |
| "kl": 0.0113983154296875, |
| "learning_rate": 8.818749999999999e-07, |
| "loss": 0.006917888764292002, |
| "reward": 2.355333089828491, |
| "reward_std": 0.26609520614147186, |
| "rewards/GDino": 0.7847018539905548, |
| "rewards/GIT": 0.5574060827493668, |
| "rewards/HPSv2": 0.2811260223388672, |
| "rewards/ORM": 0.7320991158485413, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -20.8125, |
| "step": 189 |
| }, |
| { |
| "completion_length": 43.21875, |
| "epoch": 0.21040974529346623, |
| "grad_norm": 1.6472647190093994, |
| "kl": 0.021514892578125, |
| "learning_rate": 8.812499999999999e-07, |
| "loss": 0.024224724620580673, |
| "reward": 2.3357620239257812, |
| "reward_std": 0.36145227402448654, |
| "rewards/GDino": 0.9042215049266815, |
| "rewards/GIT": 0.44460529088974, |
| "rewards/HPSv2": 0.26693153381347656, |
| "rewards/ORM": 0.720003753900528, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.125, |
| "step": 190 |
| }, |
| { |
| "completion_length": 39.84375, |
| "epoch": 0.2115171650055371, |
| "grad_norm": 0.4037325084209442, |
| "kl": 0.010467529296875, |
| "learning_rate": 8.806249999999999e-07, |
| "loss": 0.011584978085011244, |
| "reward": 2.569461226463318, |
| "reward_std": 0.31772294640541077, |
| "rewards/GDino": 0.8936654925346375, |
| "rewards/GIT": 0.6064814329147339, |
| "rewards/HPSv2": 0.28333091735839844, |
| "rewards/ORM": 0.7859834432601929, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.75, |
| "step": 191 |
| }, |
| { |
| "completion_length": 54.203125, |
| "epoch": 0.21262458471760798, |
| "grad_norm": 1.364973545074463, |
| "kl": 0.02630615234375, |
| "learning_rate": 8.799999999999999e-07, |
| "loss": -0.045025499537587166, |
| "reward": 1.7345000505447388, |
| "reward_std": 0.2504274845123291, |
| "rewards/GDino": 0.6147945821285248, |
| "rewards/GIT": 0.4107672870159149, |
| "rewards/HPSv2": 0.24933433532714844, |
| "rewards/ORM": 0.45960381627082825, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -20.4375, |
| "step": 192 |
| }, |
| { |
| "completion_length": 41.25, |
| "epoch": 0.21373200442967885, |
| "grad_norm": 0.4461517333984375, |
| "kl": 0.0135498046875, |
| "learning_rate": 8.793749999999999e-07, |
| "loss": -0.006084040272980928, |
| "reward": 1.8086916208267212, |
| "reward_std": 0.4172170013189316, |
| "rewards/GDino": 0.6766185760498047, |
| "rewards/GIT": 0.20515873283147812, |
| "rewards/HPSv2": 0.2862892150878906, |
| "rewards/ORM": 0.640625, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.875, |
| "step": 193 |
| }, |
| { |
| "completion_length": 46.046875, |
| "epoch": 0.21483942414174972, |
| "grad_norm": 0.6739068031311035, |
| "kl": 0.00982666015625, |
| "learning_rate": 8.7875e-07, |
| "loss": 0.005553322844207287, |
| "reward": 2.0827959775924683, |
| "reward_std": 0.38654056191444397, |
| "rewards/GDino": 0.8122715353965759, |
| "rewards/GIT": 0.5369465202093124, |
| "rewards/HPSv2": 0.2711658477783203, |
| "rewards/ORM": 0.46241210401058197, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -20.875, |
| "step": 194 |
| }, |
| { |
| "completion_length": 41.984375, |
| "epoch": 0.2159468438538206, |
| "grad_norm": 0.5966407060623169, |
| "kl": 0.01617431640625, |
| "learning_rate": 8.78125e-07, |
| "loss": 0.006598036969080567, |
| "reward": 2.0439035892486572, |
| "reward_std": 0.28816351294517517, |
| "rewards/GDino": 0.7164062261581421, |
| "rewards/GIT": 0.583609938621521, |
| "rewards/HPSv2": 0.27428436279296875, |
| "rewards/ORM": 0.46960312128067017, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.5, |
| "step": 195 |
| }, |
| { |
| "completion_length": 52.09375, |
| "epoch": 0.21705426356589147, |
| "grad_norm": 0.4805348217487335, |
| "kl": 0.007720947265625, |
| "learning_rate": 8.774999999999999e-07, |
| "loss": -0.005165549926459789, |
| "reward": 2.242133378982544, |
| "reward_std": 0.2688770145177841, |
| "rewards/GDino": 0.7020833194255829, |
| "rewards/GIT": 0.539276048541069, |
| "rewards/HPSv2": 0.2836284637451172, |
| "rewards/ORM": 0.7171455323696136, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.625, |
| "step": 196 |
| }, |
| { |
| "completion_length": 43.796875, |
| "epoch": 0.21816168327796234, |
| "grad_norm": 0.635823667049408, |
| "kl": 0.01513671875, |
| "learning_rate": 8.76875e-07, |
| "loss": -0.0036483434960246086, |
| "reward": 2.2711516618728638, |
| "reward_std": 0.4390978515148163, |
| "rewards/GDino": 0.6989583671092987, |
| "rewards/GIT": 0.5045495182275772, |
| "rewards/HPSv2": 0.25634765625, |
| "rewards/ORM": 0.8112961947917938, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -20.875, |
| "step": 197 |
| }, |
| { |
| "completion_length": 42.03125, |
| "epoch": 0.21926910299003322, |
| "grad_norm": 1.2338166236877441, |
| "kl": 0.012359619140625, |
| "learning_rate": 8.7625e-07, |
| "loss": 0.007895001443102956, |
| "reward": 2.048970103263855, |
| "reward_std": 0.40481944382190704, |
| "rewards/GDino": 0.8076697587966919, |
| "rewards/GIT": 0.5579419136047363, |
| "rewards/HPSv2": 0.2553253173828125, |
| "rewards/ORM": 0.42803309112787247, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.875, |
| "step": 198 |
| }, |
| { |
| "completion_length": 45.390625, |
| "epoch": 0.2203765227021041, |
| "grad_norm": 0.5174763798713684, |
| "kl": 0.01416015625, |
| "learning_rate": 8.75625e-07, |
| "loss": 0.016770444810390472, |
| "reward": 2.5844353437423706, |
| "reward_std": 0.2750231549143791, |
| "rewards/GDino": 0.8541243076324463, |
| "rewards/GIT": 0.6642543226480484, |
| "rewards/HPSv2": 0.25846099853515625, |
| "rewards/ORM": 0.8075956404209137, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.875, |
| "step": 199 |
| }, |
| { |
| "completion_length": 36.875, |
| "epoch": 0.22148394241417496, |
| "grad_norm": 0.509035587310791, |
| "kl": 0.01654052734375, |
| "learning_rate": 8.75e-07, |
| "loss": 0.006852276623249054, |
| "reward": 2.065316915512085, |
| "reward_std": 0.3316378742456436, |
| "rewards/GDino": 0.7682685256004333, |
| "rewards/GIT": 0.2910846248269081, |
| "rewards/HPSv2": 0.2672557830810547, |
| "rewards/ORM": 0.7387078404426575, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.625, |
| "step": 200 |
| }, |
| { |
| "completion_length": 42.046875, |
| "epoch": 0.22259136212624583, |
| "grad_norm": 0.5931203961372375, |
| "kl": 0.012725830078125, |
| "learning_rate": 8.74375e-07, |
| "loss": 0.006003358401358128, |
| "reward": 2.1099300384521484, |
| "reward_std": 0.5154563784599304, |
| "rewards/GDino": 0.7727163732051849, |
| "rewards/GIT": 0.3842439502477646, |
| "rewards/HPSv2": 0.2718486785888672, |
| "rewards/ORM": 0.6811210215091705, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.4375, |
| "step": 201 |
| }, |
| { |
| "completion_length": 44.171875, |
| "epoch": 0.22369878183831673, |
| "grad_norm": 0.4835968613624573, |
| "kl": 0.0213623046875, |
| "learning_rate": 8.7375e-07, |
| "loss": 0.00394545728340745, |
| "reward": 2.427845597267151, |
| "reward_std": 0.36134691536426544, |
| "rewards/GDino": 0.8869982063770294, |
| "rewards/GIT": 0.5014592558145523, |
| "rewards/HPSv2": 0.28162574768066406, |
| "rewards/ORM": 0.7577625215053558, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.8125, |
| "step": 202 |
| }, |
| { |
| "completion_length": 36.453125, |
| "epoch": 0.2248062015503876, |
| "grad_norm": 0.46910953521728516, |
| "kl": 0.0118408203125, |
| "learning_rate": 8.73125e-07, |
| "loss": 0.002786251250654459, |
| "reward": 2.49591064453125, |
| "reward_std": 0.31263355910778046, |
| "rewards/GDino": 0.8656250238418579, |
| "rewards/GIT": 0.7141720354557037, |
| "rewards/HPSv2": 0.27200889587402344, |
| "rewards/ORM": 0.6441047191619873, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.875, |
| "step": 203 |
| }, |
| { |
| "completion_length": 40.125, |
| "epoch": 0.22591362126245848, |
| "grad_norm": 0.467896968126297, |
| "kl": 0.02435302734375, |
| "learning_rate": 8.725e-07, |
| "loss": -0.011271225987002254, |
| "reward": 2.057776093482971, |
| "reward_std": 0.1546005792915821, |
| "rewards/GDino": 0.7309895753860474, |
| "rewards/GIT": 0.4367304742336273, |
| "rewards/HPSv2": 0.27594757080078125, |
| "rewards/ORM": 0.6141084581613541, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.5, |
| "step": 204 |
| }, |
| { |
| "completion_length": 33.484375, |
| "epoch": 0.22702104097452935, |
| "grad_norm": 0.7759230136871338, |
| "kl": 0.01708984375, |
| "learning_rate": 8.718749999999999e-07, |
| "loss": 0.027633000165224075, |
| "reward": 2.2592638731002808, |
| "reward_std": 0.3684349060058594, |
| "rewards/GDino": 0.7562500238418579, |
| "rewards/GIT": 0.5480997711420059, |
| "rewards/HPSv2": 0.26626014709472656, |
| "rewards/ORM": 0.6886538565158844, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -20.9375, |
| "step": 205 |
| }, |
| { |
| "completion_length": 45.046875, |
| "epoch": 0.22812846068660023, |
| "grad_norm": 0.4551739990711212, |
| "kl": 0.012664794921875, |
| "learning_rate": 8.712499999999999e-07, |
| "loss": 0.0007739269640296698, |
| "reward": 2.30593478679657, |
| "reward_std": 0.2534303367137909, |
| "rewards/GDino": 0.7875000238418579, |
| "rewards/GIT": 0.40002302825450897, |
| "rewards/HPSv2": 0.2788429260253906, |
| "rewards/ORM": 0.8395689129829407, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.8125, |
| "step": 206 |
| }, |
| { |
| "completion_length": 39.53125, |
| "epoch": 0.2292358803986711, |
| "grad_norm": 0.5830913782119751, |
| "kl": 0.01336669921875, |
| "learning_rate": 8.706249999999999e-07, |
| "loss": 0.01708681881427765, |
| "reward": 2.227774500846863, |
| "reward_std": 0.3704899549484253, |
| "rewards/GDino": 0.8083823919296265, |
| "rewards/GIT": 0.4898236393928528, |
| "rewards/HPSv2": 0.25934791564941406, |
| "rewards/ORM": 0.6702206134796143, |
| "self_certainty_semantic": -24.875, |
| "self_certainty_token": -21.5, |
| "step": 207 |
| }, |
| { |
| "completion_length": 38.5625, |
| "epoch": 0.23034330011074197, |
| "grad_norm": 0.3691738247871399, |
| "kl": 0.02581787109375, |
| "learning_rate": 8.699999999999999e-07, |
| "loss": 0.004591751378029585, |
| "reward": 2.358023166656494, |
| "reward_std": 0.17707649432122707, |
| "rewards/GDino": 0.8265625238418579, |
| "rewards/GIT": 0.6824776232242584, |
| "rewards/HPSv2": 0.2587127685546875, |
| "rewards/ORM": 0.5902703106403351, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.1875, |
| "step": 208 |
| }, |
| { |
| "completion_length": 40.3125, |
| "epoch": 0.23145071982281284, |
| "grad_norm": 0.9062660336494446, |
| "kl": 0.02685546875, |
| "learning_rate": 8.693749999999999e-07, |
| "loss": 0.006221079733222723, |
| "reward": 1.8361099362373352, |
| "reward_std": 0.4263295829296112, |
| "rewards/GDino": 0.7508396208286285, |
| "rewards/GIT": 0.3777461498975754, |
| "rewards/HPSv2": 0.266815185546875, |
| "rewards/ORM": 0.4407089203596115, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.5625, |
| "step": 209 |
| }, |
| { |
| "completion_length": 38.5625, |
| "epoch": 0.23255813953488372, |
| "grad_norm": 0.748408317565918, |
| "kl": 0.026611328125, |
| "learning_rate": 8.687499999999999e-07, |
| "loss": -0.002506181481294334, |
| "reward": 2.4682434797286987, |
| "reward_std": 0.2377740740776062, |
| "rewards/GDino": 0.8195167779922485, |
| "rewards/GIT": 0.46161703020334244, |
| "rewards/HPSv2": 0.27737998962402344, |
| "rewards/ORM": 0.9097296893596649, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.0625, |
| "step": 210 |
| }, |
| { |
| "completion_length": 41.796875, |
| "epoch": 0.2336655592469546, |
| "grad_norm": 1.765519142150879, |
| "kl": 0.02984619140625, |
| "learning_rate": 8.681249999999999e-07, |
| "loss": -0.0039495378732681274, |
| "reward": 1.913986623287201, |
| "reward_std": 0.5612352639436722, |
| "rewards/GDino": 0.6622395515441895, |
| "rewards/GIT": 0.43646256625652313, |
| "rewards/HPSv2": 0.2698841094970703, |
| "rewards/ORM": 0.545400395989418, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.5, |
| "step": 211 |
| }, |
| { |
| "completion_length": 34.96875, |
| "epoch": 0.23477297895902546, |
| "grad_norm": 0.6982265710830688, |
| "kl": 0.0120849609375, |
| "learning_rate": 8.675000000000001e-07, |
| "loss": -0.029174044728279114, |
| "reward": 1.9864473342895508, |
| "reward_std": 0.4800384193658829, |
| "rewards/GDino": 0.7475498020648956, |
| "rewards/GIT": 0.22028353065252304, |
| "rewards/HPSv2": 0.2693653106689453, |
| "rewards/ORM": 0.7492486536502838, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.5625, |
| "step": 212 |
| }, |
| { |
| "completion_length": 43.296875, |
| "epoch": 0.23588039867109634, |
| "grad_norm": 0.5722474455833435, |
| "kl": 0.016571044921875, |
| "learning_rate": 8.66875e-07, |
| "loss": -0.0017778393812477589, |
| "reward": 2.3667664527893066, |
| "reward_std": 0.40382666885852814, |
| "rewards/GDino": 0.8726562261581421, |
| "rewards/GIT": 0.6178667545318604, |
| "rewards/HPSv2": 0.2806682586669922, |
| "rewards/ORM": 0.5955753028392792, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.5625, |
| "step": 213 |
| }, |
| { |
| "completion_length": 39.46875, |
| "epoch": 0.2369878183831672, |
| "grad_norm": 0.4423932433128357, |
| "kl": 0.02313232421875, |
| "learning_rate": 8.6625e-07, |
| "loss": 0.020447181537747383, |
| "reward": 1.8110211491584778, |
| "reward_std": 0.3642263114452362, |
| "rewards/GDino": 0.6858590543270111, |
| "rewards/GIT": 0.2922380678355694, |
| "rewards/HPSv2": 0.2856483459472656, |
| "rewards/ORM": 0.5472756624221802, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -20.5625, |
| "step": 214 |
| }, |
| { |
| "completion_length": 35.453125, |
| "epoch": 0.23809523809523808, |
| "grad_norm": 0.525648295879364, |
| "kl": 0.02679443359375, |
| "learning_rate": 8.65625e-07, |
| "loss": -0.004891848191618919, |
| "reward": 2.4582409858703613, |
| "reward_std": 0.25197018682956696, |
| "rewards/GDino": 0.8230374455451965, |
| "rewards/GIT": 0.6004615277051926, |
| "rewards/HPSv2": 0.262908935546875, |
| "rewards/ORM": 0.7718330323696136, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.125, |
| "step": 215 |
| }, |
| { |
| "completion_length": 35.09375, |
| "epoch": 0.23920265780730898, |
| "grad_norm": 2.2020962238311768, |
| "kl": 0.02142333984375, |
| "learning_rate": 8.65e-07, |
| "loss": 0.009721468668431044, |
| "reward": 2.365583837032318, |
| "reward_std": 0.21866833791136742, |
| "rewards/GDino": 0.8238094449043274, |
| "rewards/GIT": 0.441422201693058, |
| "rewards/HPSv2": 0.2851238250732422, |
| "rewards/ORM": 0.8152283430099487, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.5, |
| "step": 216 |
| }, |
| { |
| "completion_length": 54.84375, |
| "epoch": 0.24031007751937986, |
| "grad_norm": 0.45819738507270813, |
| "kl": 0.012054443359375, |
| "learning_rate": 8.64375e-07, |
| "loss": -0.02724960818886757, |
| "reward": 2.1141871213912964, |
| "reward_std": 0.31379085779190063, |
| "rewards/GDino": 0.7791666686534882, |
| "rewards/GIT": 0.5268686562776566, |
| "rewards/HPSv2": 0.25792694091796875, |
| "rewards/ORM": 0.5502248406410217, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.0625, |
| "step": 217 |
| }, |
| { |
| "completion_length": 40.4375, |
| "epoch": 0.24141749723145073, |
| "grad_norm": 0.5628157258033752, |
| "kl": 0.0264892578125, |
| "learning_rate": 8.6375e-07, |
| "loss": -0.007330425316467881, |
| "reward": 1.6426080465316772, |
| "reward_std": 0.24193137884140015, |
| "rewards/GDino": 0.6079951226711273, |
| "rewards/GIT": 0.3009786158800125, |
| "rewards/HPSv2": 0.2601509094238281, |
| "rewards/ORM": 0.47348344326019287, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.125, |
| "step": 218 |
| }, |
| { |
| "completion_length": 34.625, |
| "epoch": 0.2425249169435216, |
| "grad_norm": 0.4862461984157562, |
| "kl": 0.0218505859375, |
| "learning_rate": 8.63125e-07, |
| "loss": 0.014741807244718075, |
| "reward": 2.004266083240509, |
| "reward_std": 0.32091251015663147, |
| "rewards/GDino": 0.6978627145290375, |
| "rewards/GIT": 0.45149359107017517, |
| "rewards/HPSv2": 0.27678489685058594, |
| "rewards/ORM": 0.578125, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.4375, |
| "step": 219 |
| }, |
| { |
| "completion_length": 39.875, |
| "epoch": 0.24363233665559247, |
| "grad_norm": 1.1945821046829224, |
| "kl": 0.011016845703125, |
| "learning_rate": 8.625e-07, |
| "loss": 0.02643737755715847, |
| "reward": 2.016072630882263, |
| "reward_std": 0.323383167386055, |
| "rewards/GDino": 0.8033246994018555, |
| "rewards/GIT": 0.5735934674739838, |
| "rewards/HPSv2": 0.25634193420410156, |
| "rewards/ORM": 0.3828125, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.25, |
| "step": 220 |
| }, |
| { |
| "completion_length": 35.546875, |
| "epoch": 0.24473975636766335, |
| "grad_norm": 0.9685459136962891, |
| "kl": 0.01617431640625, |
| "learning_rate": 8.618749999999999e-07, |
| "loss": -0.01451091282069683, |
| "reward": 1.7261735200881958, |
| "reward_std": 0.4147494286298752, |
| "rewards/GDino": 0.6085260510444641, |
| "rewards/GIT": 0.2463730275630951, |
| "rewards/HPSv2": 0.2583179473876953, |
| "rewards/ORM": 0.6129564046859741, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.5, |
| "step": 221 |
| }, |
| { |
| "completion_length": 34.890625, |
| "epoch": 0.24584717607973422, |
| "grad_norm": 0.5475857257843018, |
| "kl": 0.022491455078125, |
| "learning_rate": 8.612499999999999e-07, |
| "loss": 0.033416807651519775, |
| "reward": 1.9957242012023926, |
| "reward_std": 0.5047579407691956, |
| "rewards/GDino": 0.658912181854248, |
| "rewards/GIT": 0.4291614145040512, |
| "rewards/HPSv2": 0.271759033203125, |
| "rewards/ORM": 0.6358915567398071, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.25, |
| "step": 222 |
| }, |
| { |
| "completion_length": 36.515625, |
| "epoch": 0.2469545957918051, |
| "grad_norm": 0.4916452169418335, |
| "kl": 0.0184326171875, |
| "learning_rate": 8.606249999999999e-07, |
| "loss": -0.00917936791665852, |
| "reward": 2.0419200658798218, |
| "reward_std": 0.5100755244493484, |
| "rewards/GDino": 0.729717344045639, |
| "rewards/GIT": 0.21226511895656586, |
| "rewards/HPSv2": 0.2703380584716797, |
| "rewards/ORM": 0.8295995891094208, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.75, |
| "step": 223 |
| }, |
| { |
| "completion_length": 30.46875, |
| "epoch": 0.24806201550387597, |
| "grad_norm": 0.5015460252761841, |
| "kl": 0.0233154296875, |
| "learning_rate": 8.599999999999999e-07, |
| "loss": -0.0018524383194744587, |
| "reward": 2.36031711101532, |
| "reward_std": 0.19695958495140076, |
| "rewards/GDino": 0.820055365562439, |
| "rewards/GIT": 0.45940345525741577, |
| "rewards/HPSv2": 0.2683582305908203, |
| "rewards/ORM": 0.8125, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.1875, |
| "step": 224 |
| }, |
| { |
| "completion_length": 41.0, |
| "epoch": 0.24916943521594684, |
| "grad_norm": 0.5461787581443787, |
| "kl": 0.025390625, |
| "learning_rate": 8.593749999999999e-07, |
| "loss": -0.009375374997034669, |
| "reward": 2.1164638996124268, |
| "reward_std": 0.4978756159543991, |
| "rewards/GDino": 0.7708333432674408, |
| "rewards/GIT": 0.5871168673038483, |
| "rewards/HPSv2": 0.237884521484375, |
| "rewards/ORM": 0.5206292420625687, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.1875, |
| "step": 225 |
| }, |
| { |
| "completion_length": 30.96875, |
| "epoch": 0.2502768549280177, |
| "grad_norm": 0.46695247292518616, |
| "kl": 0.016510009765625, |
| "learning_rate": 8.587499999999999e-07, |
| "loss": -0.0015379427932202816, |
| "reward": 2.3315041065216064, |
| "reward_std": 0.35161057114601135, |
| "rewards/GDino": 0.8348958492279053, |
| "rewards/GIT": 0.41105780750513077, |
| "rewards/HPSv2": 0.271575927734375, |
| "rewards/ORM": 0.8139745891094208, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.4375, |
| "step": 226 |
| }, |
| { |
| "completion_length": 38.703125, |
| "epoch": 0.2513842746400886, |
| "grad_norm": 0.5733257532119751, |
| "kl": 0.01409912109375, |
| "learning_rate": 8.581249999999999e-07, |
| "loss": -0.003186628222465515, |
| "reward": 1.9469320178031921, |
| "reward_std": 0.47824153304100037, |
| "rewards/GDino": 0.7051927149295807, |
| "rewards/GIT": 0.42935653030872345, |
| "rewards/HPSv2": 0.2548370361328125, |
| "rewards/ORM": 0.5575457066297531, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.4375, |
| "step": 227 |
| }, |
| { |
| "completion_length": 43.234375, |
| "epoch": 0.25249169435215946, |
| "grad_norm": 0.40495043992996216, |
| "kl": 0.016357421875, |
| "learning_rate": 8.575e-07, |
| "loss": -0.00824080128222704, |
| "reward": 1.9755643606185913, |
| "reward_std": 0.3469914644956589, |
| "rewards/GDino": 0.8501654267311096, |
| "rewards/GIT": 0.19020378589630127, |
| "rewards/HPSv2": 0.2816734313964844, |
| "rewards/ORM": 0.6535216271877289, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.5, |
| "step": 228 |
| }, |
| { |
| "completion_length": 40.453125, |
| "epoch": 0.25359911406423036, |
| "grad_norm": 0.6345773935317993, |
| "kl": 0.019287109375, |
| "learning_rate": 8.568750000000001e-07, |
| "loss": 0.0025403383187949657, |
| "reward": 2.108458161354065, |
| "reward_std": 0.28643767535686493, |
| "rewards/GDino": 0.6715624928474426, |
| "rewards/GIT": 0.3562132343649864, |
| "rewards/HPSv2": 0.2619743347167969, |
| "rewards/ORM": 0.818708062171936, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.0625, |
| "step": 229 |
| }, |
| { |
| "completion_length": 38.8125, |
| "epoch": 0.2547065337763012, |
| "grad_norm": 0.5043112635612488, |
| "kl": 0.01556396484375, |
| "learning_rate": 8.5625e-07, |
| "loss": 0.009474070742726326, |
| "reward": 2.0315651893615723, |
| "reward_std": 0.4527067393064499, |
| "rewards/GDino": 0.7280016541481018, |
| "rewards/GIT": 0.34433120489120483, |
| "rewards/HPSv2": 0.2717323303222656, |
| "rewards/ORM": 0.6875, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0625, |
| "step": 230 |
| }, |
| { |
| "completion_length": 39.75, |
| "epoch": 0.2558139534883721, |
| "grad_norm": 0.7560561299324036, |
| "kl": 0.0228271484375, |
| "learning_rate": 8.55625e-07, |
| "loss": 0.03068174561485648, |
| "reward": 1.8725414276123047, |
| "reward_std": 0.37937965989112854, |
| "rewards/GDino": 0.7148861289024353, |
| "rewards/GIT": 0.35717466473579407, |
| "rewards/HPSv2": 0.258209228515625, |
| "rewards/ORM": 0.5422714352607727, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.0625, |
| "step": 231 |
| }, |
| { |
| "completion_length": 39.296875, |
| "epoch": 0.25692137320044295, |
| "grad_norm": 1.4232704639434814, |
| "kl": 0.013641357421875, |
| "learning_rate": 8.55e-07, |
| "loss": 0.007876838091760874, |
| "reward": 1.9952596426010132, |
| "reward_std": 0.4007260948419571, |
| "rewards/GDino": 0.7144716680049896, |
| "rewards/GIT": 0.4604896456003189, |
| "rewards/HPSv2": 0.25359535217285156, |
| "rewards/ORM": 0.5667029619216919, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.0, |
| "step": 232 |
| }, |
| { |
| "completion_length": 52.671875, |
| "epoch": 0.25802879291251385, |
| "grad_norm": 0.553196907043457, |
| "kl": 0.01513671875, |
| "learning_rate": 8.54375e-07, |
| "loss": -0.002937940414994955, |
| "reward": 2.1501262187957764, |
| "reward_std": 0.3424035310745239, |
| "rewards/GDino": 0.7523437142372131, |
| "rewards/GIT": 0.3163189962506294, |
| "rewards/HPSv2": 0.2654838562011719, |
| "rewards/ORM": 0.8159796893596649, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.4375, |
| "step": 233 |
| }, |
| { |
| "completion_length": 35.40625, |
| "epoch": 0.2591362126245847, |
| "grad_norm": 4.139526844024658, |
| "kl": 0.033203125, |
| "learning_rate": 8.5375e-07, |
| "loss": 0.007489941082894802, |
| "reward": 2.0956884622573853, |
| "reward_std": 0.25674766674637794, |
| "rewards/GDino": 0.8046875, |
| "rewards/GIT": 0.30966047942638397, |
| "rewards/HPSv2": 0.2782154083251953, |
| "rewards/ORM": 0.703125, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -20.9375, |
| "step": 234 |
| }, |
| { |
| "completion_length": 39.875, |
| "epoch": 0.2602436323366556, |
| "grad_norm": 0.509060263633728, |
| "kl": 0.0189208984375, |
| "learning_rate": 8.53125e-07, |
| "loss": 0.004633883247151971, |
| "reward": 1.9408237934112549, |
| "reward_std": 0.4254070520401001, |
| "rewards/GDino": 0.7614295780658722, |
| "rewards/GIT": 0.3818518742918968, |
| "rewards/HPSv2": 0.27335357666015625, |
| "rewards/ORM": 0.5241888463497162, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.625, |
| "step": 235 |
| }, |
| { |
| "completion_length": 36.375, |
| "epoch": 0.26135105204872644, |
| "grad_norm": 0.4405996799468994, |
| "kl": 0.0157470703125, |
| "learning_rate": 8.525e-07, |
| "loss": -0.010419015074148774, |
| "reward": 1.502617061138153, |
| "reward_std": 0.4193390905857086, |
| "rewards/GDino": 0.7078101933002472, |
| "rewards/GIT": 0.2298043891787529, |
| "rewards/HPSv2": 0.25250244140625, |
| "rewards/ORM": 0.3125000149011612, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.4375, |
| "step": 236 |
| }, |
| { |
| "completion_length": 39.0, |
| "epoch": 0.26245847176079734, |
| "grad_norm": 0.505985677242279, |
| "kl": 0.01690673828125, |
| "learning_rate": 8.51875e-07, |
| "loss": -0.016914513893425465, |
| "reward": 2.266430974006653, |
| "reward_std": 0.4797457307577133, |
| "rewards/GDino": 0.7984375059604645, |
| "rewards/GIT": 0.49608829617500305, |
| "rewards/HPSv2": 0.27498817443847656, |
| "rewards/ORM": 0.6969169676303864, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.625, |
| "step": 237 |
| }, |
| { |
| "completion_length": 35.3125, |
| "epoch": 0.26356589147286824, |
| "grad_norm": 0.484314888715744, |
| "kl": 0.0130615234375, |
| "learning_rate": 8.512499999999999e-07, |
| "loss": 0.030976343899965286, |
| "reward": 1.991338849067688, |
| "reward_std": 0.2668539509177208, |
| "rewards/GDino": 0.8419585227966309, |
| "rewards/GIT": 0.44607045501470566, |
| "rewards/HPSv2": 0.26688385009765625, |
| "rewards/ORM": 0.436426043510437, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -20.125, |
| "step": 238 |
| }, |
| { |
| "completion_length": 44.828125, |
| "epoch": 0.2646733111849391, |
| "grad_norm": 0.7890397310256958, |
| "kl": 0.01763916015625, |
| "learning_rate": 8.506249999999999e-07, |
| "loss": -0.0025457546580582857, |
| "reward": 2.0841389894485474, |
| "reward_std": 0.19053151458501816, |
| "rewards/GDino": 0.6302083730697632, |
| "rewards/GIT": 0.5595559030771255, |
| "rewards/HPSv2": 0.2693748474121094, |
| "rewards/ORM": 0.625, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -20.4375, |
| "step": 239 |
| }, |
| { |
| "completion_length": 44.859375, |
| "epoch": 0.26578073089701, |
| "grad_norm": 0.8127689361572266, |
| "kl": 0.022491455078125, |
| "learning_rate": 8.499999999999999e-07, |
| "loss": 0.0033242302015423775, |
| "reward": 2.0860930681228638, |
| "reward_std": 0.22588412836194038, |
| "rewards/GDino": 0.7918341755867004, |
| "rewards/GIT": 0.4664677008986473, |
| "rewards/HPSv2": 0.2652912139892578, |
| "rewards/ORM": 0.5625, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.375, |
| "step": 240 |
| }, |
| { |
| "completion_length": 38.6875, |
| "epoch": 0.26688815060908083, |
| "grad_norm": 0.7459165453910828, |
| "kl": 0.0216064453125, |
| "learning_rate": 8.493749999999999e-07, |
| "loss": -0.018312662839889526, |
| "reward": 2.1917264461517334, |
| "reward_std": 0.19818688184022903, |
| "rewards/GDino": 0.765625, |
| "rewards/GIT": 0.5940662920475006, |
| "rewards/HPSv2": 0.2635478973388672, |
| "rewards/ORM": 0.5684872269630432, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.0625, |
| "step": 241 |
| }, |
| { |
| "completion_length": 40.1875, |
| "epoch": 0.26799557032115173, |
| "grad_norm": 0.528344988822937, |
| "kl": 0.013427734375, |
| "learning_rate": 8.487499999999999e-07, |
| "loss": 0.0015740576200187206, |
| "reward": 1.820701777935028, |
| "reward_std": 0.2884362041950226, |
| "rewards/GDino": 0.7261866927146912, |
| "rewards/GIT": 0.32748638093471527, |
| "rewards/HPSv2": 0.2674293518066406, |
| "rewards/ORM": 0.4995993375778198, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.75, |
| "step": 242 |
| }, |
| { |
| "completion_length": 35.640625, |
| "epoch": 0.2691029900332226, |
| "grad_norm": 0.6940242648124695, |
| "kl": 0.0177001953125, |
| "learning_rate": 8.481249999999999e-07, |
| "loss": -0.0034725347068160772, |
| "reward": 1.5581411123275757, |
| "reward_std": 0.4635423719882965, |
| "rewards/GDino": 0.6132907867431641, |
| "rewards/GIT": 0.22624823451042175, |
| "rewards/HPSv2": 0.2540550231933594, |
| "rewards/ORM": 0.4645470678806305, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -20.625, |
| "step": 243 |
| }, |
| { |
| "completion_length": 41.46875, |
| "epoch": 0.2702104097452935, |
| "grad_norm": 0.5053713321685791, |
| "kl": 0.0189208984375, |
| "learning_rate": 8.475e-07, |
| "loss": -0.0003409823402762413, |
| "reward": 1.8292673826217651, |
| "reward_std": 0.3081812709569931, |
| "rewards/GDino": 0.7110464870929718, |
| "rewards/GIT": 0.18649065494537354, |
| "rewards/HPSv2": 0.2602558135986328, |
| "rewards/ORM": 0.6714743673801422, |
| "self_certainty_semantic": -24.875, |
| "self_certainty_token": -21.25, |
| "step": 244 |
| }, |
| { |
| "completion_length": 36.484375, |
| "epoch": 0.2713178294573643, |
| "grad_norm": 0.6690962910652161, |
| "kl": 0.0186767578125, |
| "learning_rate": 8.46875e-07, |
| "loss": -0.020096718333661556, |
| "reward": 1.7160680294036865, |
| "reward_std": 0.4093341752886772, |
| "rewards/GDino": 0.7332078814506531, |
| "rewards/GIT": 0.2718503773212433, |
| "rewards/HPSv2": 0.26757240295410156, |
| "rewards/ORM": 0.44343726336956024, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.9375, |
| "step": 245 |
| }, |
| { |
| "completion_length": 42.96875, |
| "epoch": 0.2724252491694352, |
| "grad_norm": 0.7113038897514343, |
| "kl": 0.0272216796875, |
| "learning_rate": 8.462499999999999e-07, |
| "loss": 0.021475711837410927, |
| "reward": 2.2206690311431885, |
| "reward_std": 0.2255793958902359, |
| "rewards/GDino": 0.8505983948707581, |
| "rewards/GIT": 0.5832930952310562, |
| "rewards/HPSv2": 0.24771499633789062, |
| "rewards/ORM": 0.5390624701976776, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.0625, |
| "step": 246 |
| }, |
| { |
| "completion_length": 39.3125, |
| "epoch": 0.27353266888150607, |
| "grad_norm": 1.8593120574951172, |
| "kl": 0.0225830078125, |
| "learning_rate": 8.45625e-07, |
| "loss": -0.00028415233828127384, |
| "reward": 2.0454858541488647, |
| "reward_std": 0.3580029234290123, |
| "rewards/GDino": 0.7631928622722626, |
| "rewards/GIT": 0.3862551599740982, |
| "rewards/HPSv2": 0.2628746032714844, |
| "rewards/ORM": 0.6331632137298584, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -20.9375, |
| "step": 247 |
| }, |
| { |
| "completion_length": 36.125, |
| "epoch": 0.27464008859357697, |
| "grad_norm": 0.592446506023407, |
| "kl": 0.02227783203125, |
| "learning_rate": 8.45e-07, |
| "loss": 0.02284400351345539, |
| "reward": 1.5269662737846375, |
| "reward_std": 0.4807903617620468, |
| "rewards/GDino": 0.6762161254882812, |
| "rewards/GIT": 0.23257604241371155, |
| "rewards/HPSv2": 0.28420257568359375, |
| "rewards/ORM": 0.3339715301990509, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.3125, |
| "step": 248 |
| }, |
| { |
| "completion_length": 38.28125, |
| "epoch": 0.2757475083056478, |
| "grad_norm": 0.6646723747253418, |
| "kl": 0.01947021484375, |
| "learning_rate": 8.44375e-07, |
| "loss": -0.0015255426988005638, |
| "reward": 2.369264245033264, |
| "reward_std": 0.45240701735019684, |
| "rewards/GDino": 0.803423672914505, |
| "rewards/GIT": 0.6715860366821289, |
| "rewards/HPSv2": 0.27072906494140625, |
| "rewards/ORM": 0.6235254108905792, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.5, |
| "step": 249 |
| }, |
| { |
| "completion_length": 39.859375, |
| "epoch": 0.2768549280177187, |
| "grad_norm": 0.5067113637924194, |
| "kl": 0.02447509765625, |
| "learning_rate": 8.4375e-07, |
| "loss": 0.010950385592877865, |
| "reward": 2.4666439294815063, |
| "reward_std": 0.2641718164086342, |
| "rewards/GDino": 0.8098958134651184, |
| "rewards/GIT": 0.8221855163574219, |
| "rewards/HPSv2": 0.2626457214355469, |
| "rewards/ORM": 0.571916937828064, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.1875, |
| "step": 250 |
| }, |
| { |
| "completion_length": 48.34375, |
| "epoch": 0.2779623477297896, |
| "grad_norm": 0.5123341679573059, |
| "kl": 0.014495849609375, |
| "learning_rate": 8.43125e-07, |
| "loss": -0.007521981373429298, |
| "reward": 2.471664547920227, |
| "reward_std": 0.3560677170753479, |
| "rewards/GDino": 0.870312511920929, |
| "rewards/GIT": 0.761746883392334, |
| "rewards/HPSv2": 0.2522430419921875, |
| "rewards/ORM": 0.587362140417099, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -20.8125, |
| "step": 251 |
| }, |
| { |
| "completion_length": 39.125, |
| "epoch": 0.27906976744186046, |
| "grad_norm": 0.5467274785041809, |
| "kl": 0.0211181640625, |
| "learning_rate": 8.425e-07, |
| "loss": 0.0022668561432510614, |
| "reward": 1.6477148532867432, |
| "reward_std": 0.5759541094303131, |
| "rewards/GDino": 0.6465478837490082, |
| "rewards/GIT": 0.18030283600091934, |
| "rewards/HPSv2": 0.2717132568359375, |
| "rewards/ORM": 0.549150824546814, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.25, |
| "step": 252 |
| }, |
| { |
| "completion_length": 29.4375, |
| "epoch": 0.28017718715393136, |
| "grad_norm": 0.4424716532230377, |
| "kl": 0.01275634765625, |
| "learning_rate": 8.41875e-07, |
| "loss": 0.00682451156899333, |
| "reward": 2.520264148712158, |
| "reward_std": 0.2268836908042431, |
| "rewards/GDino": 0.8885416686534882, |
| "rewards/GIT": 0.6353528201580048, |
| "rewards/HPSv2": 0.2576618194580078, |
| "rewards/ORM": 0.7387077957391739, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -21.375, |
| "step": 253 |
| }, |
| { |
| "completion_length": 33.453125, |
| "epoch": 0.2812846068660022, |
| "grad_norm": 1.276292324066162, |
| "kl": 0.0240478515625, |
| "learning_rate": 8.4125e-07, |
| "loss": 0.021866907365620136, |
| "reward": 2.1942864656448364, |
| "reward_std": 0.2500537782907486, |
| "rewards/GDino": 0.7828124761581421, |
| "rewards/GIT": 0.6722497642040253, |
| "rewards/HPSv2": 0.2669944763183594, |
| "rewards/ORM": 0.4722296893596649, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.6875, |
| "step": 254 |
| }, |
| { |
| "completion_length": 43.296875, |
| "epoch": 0.2823920265780731, |
| "grad_norm": 0.6294586658477783, |
| "kl": 0.02001953125, |
| "learning_rate": 8.406249999999999e-07, |
| "loss": -0.013118594884872437, |
| "reward": 2.0315213203430176, |
| "reward_std": 0.26577161997556686, |
| "rewards/GDino": 0.6790449321269989, |
| "rewards/GIT": 0.3302696421742439, |
| "rewards/HPSv2": 0.2811431884765625, |
| "rewards/ORM": 0.7410635948181152, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.0625, |
| "step": 255 |
| }, |
| { |
| "completion_length": 38.234375, |
| "epoch": 0.28349944629014395, |
| "grad_norm": 0.47897863388061523, |
| "kl": 0.02569580078125, |
| "learning_rate": 8.399999999999999e-07, |
| "loss": 0.007225595414638519, |
| "reward": 1.8496492505073547, |
| "reward_std": 0.4634602516889572, |
| "rewards/GDino": 0.6612989604473114, |
| "rewards/GIT": 0.3624468967318535, |
| "rewards/HPSv2": 0.2617988586425781, |
| "rewards/ORM": 0.564104437828064, |
| "self_certainty_semantic": -24.875, |
| "self_certainty_token": -21.125, |
| "step": 256 |
| }, |
| { |
| "completion_length": 34.90625, |
| "epoch": 0.28460686600221485, |
| "grad_norm": 1.0091742277145386, |
| "kl": 0.011627197265625, |
| "learning_rate": 8.393749999999999e-07, |
| "loss": -0.003027831669896841, |
| "reward": 1.818339228630066, |
| "reward_std": 0.34789541363716125, |
| "rewards/GDino": 0.68548583984375, |
| "rewards/GIT": 0.3273261487483978, |
| "rewards/HPSv2": 0.26115989685058594, |
| "rewards/ORM": 0.5443674623966217, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.875, |
| "step": 257 |
| }, |
| { |
| "completion_length": 42.328125, |
| "epoch": 0.2857142857142857, |
| "grad_norm": 0.9769588708877563, |
| "kl": 0.01727294921875, |
| "learning_rate": 8.387499999999999e-07, |
| "loss": -0.027014064602553844, |
| "reward": 1.6609314680099487, |
| "reward_std": 0.5478866696357727, |
| "rewards/GDino": 0.5854167342185974, |
| "rewards/GIT": 0.18769004940986633, |
| "rewards/HPSv2": 0.27184104919433594, |
| "rewards/ORM": 0.6159836798906326, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.0625, |
| "step": 258 |
| }, |
| { |
| "completion_length": 34.203125, |
| "epoch": 0.2868217054263566, |
| "grad_norm": 0.4485263526439667, |
| "kl": 0.019287109375, |
| "learning_rate": 8.38125e-07, |
| "loss": 0.018385295988991857, |
| "reward": 2.5785293579101562, |
| "reward_std": 0.17483927309513092, |
| "rewards/GDino": 0.9424871504306793, |
| "rewards/GIT": 0.7919199466705322, |
| "rewards/HPSv2": 0.2565803527832031, |
| "rewards/ORM": 0.5875419676303864, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.1875, |
| "step": 259 |
| }, |
| { |
| "completion_length": 37.203125, |
| "epoch": 0.28792912513842744, |
| "grad_norm": 0.449808806180954, |
| "kl": 0.0335693359375, |
| "learning_rate": 8.375e-07, |
| "loss": 0.014663147740066051, |
| "reward": 1.8461943864822388, |
| "reward_std": 0.5021175146102905, |
| "rewards/GDino": 0.6706274747848511, |
| "rewards/GIT": 0.3253522589802742, |
| "rewards/HPSv2": 0.27150726318359375, |
| "rewards/ORM": 0.5787073373794556, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.3125, |
| "step": 260 |
| }, |
| { |
| "completion_length": 43.40625, |
| "epoch": 0.28903654485049834, |
| "grad_norm": 0.6159392595291138, |
| "kl": 0.02618408203125, |
| "learning_rate": 8.36875e-07, |
| "loss": 0.006308391690254211, |
| "reward": 2.2187058329582214, |
| "reward_std": 0.3276343122124672, |
| "rewards/GDino": 0.7517669200897217, |
| "rewards/GIT": 0.5226505398750305, |
| "rewards/HPSv2": 0.2646007537841797, |
| "rewards/ORM": 0.6796875, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.3125, |
| "step": 261 |
| }, |
| { |
| "completion_length": 40.84375, |
| "epoch": 0.2901439645625692, |
| "grad_norm": 0.4515726864337921, |
| "kl": 0.0157470703125, |
| "learning_rate": 8.3625e-07, |
| "loss": -0.025780703872442245, |
| "reward": 2.513529419898987, |
| "reward_std": 0.377413272857666, |
| "rewards/GDino": 0.8145833611488342, |
| "rewards/GIT": 0.6835201978683472, |
| "rewards/HPSv2": 0.2530097961425781, |
| "rewards/ORM": 0.7624161243438721, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.0, |
| "step": 262 |
| }, |
| { |
| "completion_length": 47.8125, |
| "epoch": 0.2912513842746401, |
| "grad_norm": 0.6491768956184387, |
| "kl": 0.01971435546875, |
| "learning_rate": 8.356249999999999e-07, |
| "loss": -0.010318214073777199, |
| "reward": 2.480243444442749, |
| "reward_std": 0.38080593943595886, |
| "rewards/GDino": 0.8481250107288361, |
| "rewards/GIT": 0.49081389605998993, |
| "rewards/HPSv2": 0.25723838806152344, |
| "rewards/ORM": 0.884066253900528, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.75, |
| "step": 263 |
| }, |
| { |
| "completion_length": 42.140625, |
| "epoch": 0.292358803986711, |
| "grad_norm": 0.635397732257843, |
| "kl": 0.027191162109375, |
| "learning_rate": 8.349999999999999e-07, |
| "loss": -0.010353719699196517, |
| "reward": 2.442339539527893, |
| "reward_std": 0.32322730123996735, |
| "rewards/GDino": 0.8117945790290833, |
| "rewards/GIT": 0.6421481668949127, |
| "rewards/HPSv2": 0.2681722640991211, |
| "rewards/ORM": 0.7202245593070984, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -20.5625, |
| "step": 264 |
| }, |
| { |
| "completion_length": 43.703125, |
| "epoch": 0.29346622369878184, |
| "grad_norm": 0.4443694055080414, |
| "kl": 0.0185546875, |
| "learning_rate": 8.34375e-07, |
| "loss": 0.02428914955817163, |
| "reward": 2.2206848859786987, |
| "reward_std": 0.4008527547121048, |
| "rewards/GDino": 0.8173159956932068, |
| "rewards/GIT": 0.43181733787059784, |
| "rewards/HPSv2": 0.27811431884765625, |
| "rewards/ORM": 0.693437248468399, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.0625, |
| "step": 265 |
| }, |
| { |
| "completion_length": 43.875, |
| "epoch": 0.29457364341085274, |
| "grad_norm": 0.5283258557319641, |
| "kl": 0.015289306640625, |
| "learning_rate": 8.3375e-07, |
| "loss": 0.019315090496093035, |
| "reward": 1.8423295617103577, |
| "reward_std": 0.337589330971241, |
| "rewards/GDino": 0.6214767098426819, |
| "rewards/GIT": 0.3616393804550171, |
| "rewards/HPSv2": 0.27019691467285156, |
| "rewards/ORM": 0.5890165567398071, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.0, |
| "step": 266 |
| }, |
| { |
| "completion_length": 42.578125, |
| "epoch": 0.2956810631229236, |
| "grad_norm": 0.5958583950996399, |
| "kl": 0.02301025390625, |
| "learning_rate": 8.33125e-07, |
| "loss": -0.010422109626233578, |
| "reward": 2.478596568107605, |
| "reward_std": 0.2979115843772888, |
| "rewards/GDino": 0.8439387381076813, |
| "rewards/GIT": 0.5261026620864868, |
| "rewards/HPSv2": 0.28636741638183594, |
| "rewards/ORM": 0.8221877217292786, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.9375, |
| "step": 267 |
| }, |
| { |
| "completion_length": 38.8125, |
| "epoch": 0.2967884828349945, |
| "grad_norm": 0.5406501889228821, |
| "kl": 0.0189208984375, |
| "learning_rate": 8.325e-07, |
| "loss": -0.011615738272666931, |
| "reward": 2.1584006547927856, |
| "reward_std": 0.3037988841533661, |
| "rewards/GDino": 0.7368161380290985, |
| "rewards/GIT": 0.3939068764448166, |
| "rewards/HPSv2": 0.2691917419433594, |
| "rewards/ORM": 0.7584857940673828, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.4375, |
| "step": 268 |
| }, |
| { |
| "completion_length": 40.375, |
| "epoch": 0.2978959025470653, |
| "grad_norm": 0.5237568616867065, |
| "kl": 0.01934814453125, |
| "learning_rate": 8.31875e-07, |
| "loss": -0.022221547085791826, |
| "reward": 2.410371422767639, |
| "reward_std": 0.32327765971422195, |
| "rewards/GDino": 0.8763707876205444, |
| "rewards/GIT": 0.6529507040977478, |
| "rewards/HPSv2": 0.25867652893066406, |
| "rewards/ORM": 0.622373417019844, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.0625, |
| "step": 269 |
| }, |
| { |
| "completion_length": 42.40625, |
| "epoch": 0.29900332225913623, |
| "grad_norm": 0.5596421957015991, |
| "kl": 0.02099609375, |
| "learning_rate": 8.3125e-07, |
| "loss": 0.0010367396753281355, |
| "reward": 1.9419864416122437, |
| "reward_std": 0.3841004818677902, |
| "rewards/GDino": 0.7421875298023224, |
| "rewards/GIT": 0.40068522095680237, |
| "rewards/HPSv2": 0.26871681213378906, |
| "rewards/ORM": 0.5303968787193298, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.625, |
| "step": 270 |
| }, |
| { |
| "completion_length": 36.515625, |
| "epoch": 0.3001107419712071, |
| "grad_norm": 0.4112670421600342, |
| "kl": 0.02191162109375, |
| "learning_rate": 8.306249999999999e-07, |
| "loss": 0.008409947156906128, |
| "reward": 1.5938773155212402, |
| "reward_std": 0.4974990785121918, |
| "rewards/GDino": 0.7035039365291595, |
| "rewards/GIT": 0.2642563134431839, |
| "rewards/HPSv2": 0.24611282348632812, |
| "rewards/ORM": 0.3800041973590851, |
| "self_certainty_semantic": -24.875, |
| "self_certainty_token": -22.3125, |
| "step": 271 |
| }, |
| { |
| "completion_length": 42.453125, |
| "epoch": 0.301218161683278, |
| "grad_norm": 1.410166621208191, |
| "kl": 0.018798828125, |
| "learning_rate": 8.299999999999999e-07, |
| "loss": -0.024008065927773714, |
| "reward": 1.7751591801643372, |
| "reward_std": 0.4845600575208664, |
| "rewards/GDino": 0.6045292317867279, |
| "rewards/GIT": 0.2201373279094696, |
| "rewards/HPSv2": 0.27045440673828125, |
| "rewards/ORM": 0.6800382286310196, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.375, |
| "step": 272 |
| }, |
| { |
| "completion_length": 34.78125, |
| "epoch": 0.3023255813953488, |
| "grad_norm": 1.887034296989441, |
| "kl": 0.01806640625, |
| "learning_rate": 8.293749999999999e-07, |
| "loss": 0.00430523045361042, |
| "reward": 1.7456142902374268, |
| "reward_std": 0.3797690123319626, |
| "rewards/GDino": 0.658672958612442, |
| "rewards/GIT": 0.3142998740077019, |
| "rewards/HPSv2": 0.2761211395263672, |
| "rewards/ORM": 0.4965203106403351, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.0625, |
| "step": 273 |
| }, |
| { |
| "completion_length": 35.328125, |
| "epoch": 0.3034330011074197, |
| "grad_norm": 0.9875066876411438, |
| "kl": 0.02935791015625, |
| "learning_rate": 8.287499999999999e-07, |
| "loss": 0.013776570558547974, |
| "reward": 2.0887337923049927, |
| "reward_std": 0.39243339002132416, |
| "rewards/GDino": 0.7666075229644775, |
| "rewards/GIT": 0.5064676105976105, |
| "rewards/HPSv2": 0.25235748291015625, |
| "rewards/ORM": 0.5633012652397156, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.125, |
| "step": 274 |
| }, |
| { |
| "completion_length": 42.015625, |
| "epoch": 0.30454042081949056, |
| "grad_norm": 0.5050070285797119, |
| "kl": 0.019317626953125, |
| "learning_rate": 8.28125e-07, |
| "loss": 0.01794680766761303, |
| "reward": 2.768193483352661, |
| "reward_std": 0.20595969259738922, |
| "rewards/GDino": 0.8856770694255829, |
| "rewards/GIT": 0.7785017788410187, |
| "rewards/HPSv2": 0.2793693542480469, |
| "rewards/ORM": 0.8246453106403351, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.875, |
| "step": 275 |
| }, |
| { |
| "completion_length": 36.234375, |
| "epoch": 0.30564784053156147, |
| "grad_norm": 0.8548435568809509, |
| "kl": 0.015167236328125, |
| "learning_rate": 8.275e-07, |
| "loss": -0.001379463355988264, |
| "reward": 2.29227876663208, |
| "reward_std": 0.3256339356303215, |
| "rewards/GDino": 0.7592664659023285, |
| "rewards/GIT": 0.4296427518129349, |
| "rewards/HPSv2": 0.27176475524902344, |
| "rewards/ORM": 0.8316046893596649, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.0625, |
| "step": 276 |
| }, |
| { |
| "completion_length": 38.875, |
| "epoch": 0.3067552602436323, |
| "grad_norm": 0.49069884419441223, |
| "kl": 0.02545166015625, |
| "learning_rate": 8.26875e-07, |
| "loss": 0.003206422086805105, |
| "reward": 2.004135310649872, |
| "reward_std": 0.4736744314432144, |
| "rewards/GDino": 0.6578125059604645, |
| "rewards/GIT": 0.5337191522121429, |
| "rewards/HPSv2": 0.2700614929199219, |
| "rewards/ORM": 0.5425421595573425, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.75, |
| "step": 277 |
| }, |
| { |
| "completion_length": 33.703125, |
| "epoch": 0.3078626799557032, |
| "grad_norm": 0.5431483387947083, |
| "kl": 0.029052734375, |
| "learning_rate": 8.2625e-07, |
| "loss": -0.01096310350112617, |
| "reward": 2.36252760887146, |
| "reward_std": 0.3281755894422531, |
| "rewards/GDino": 0.8864299952983856, |
| "rewards/GIT": 0.6353955864906311, |
| "rewards/HPSv2": 0.2625770568847656, |
| "rewards/ORM": 0.578125, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -20.875, |
| "step": 278 |
| }, |
| { |
| "completion_length": 40.84375, |
| "epoch": 0.3089700996677741, |
| "grad_norm": 0.6272196173667908, |
| "kl": 0.017791748046875, |
| "learning_rate": 8.25625e-07, |
| "loss": 0.005458365194499493, |
| "reward": 2.53033185005188, |
| "reward_std": 0.3404112756252289, |
| "rewards/GDino": 0.8331500887870789, |
| "rewards/GIT": 0.7181890606880188, |
| "rewards/HPSv2": 0.2650260925292969, |
| "rewards/ORM": 0.7139666378498077, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.25, |
| "step": 279 |
| }, |
| { |
| "completion_length": 29.625, |
| "epoch": 0.31007751937984496, |
| "grad_norm": 0.4168836176395416, |
| "kl": 0.013580322265625, |
| "learning_rate": 8.249999999999999e-07, |
| "loss": -0.020269228611141443, |
| "reward": 2.1206799745559692, |
| "reward_std": 0.36074501276016235, |
| "rewards/GDino": 0.7140624821186066, |
| "rewards/GIT": 0.5316184759140015, |
| "rewards/HPSv2": 0.24579620361328125, |
| "rewards/ORM": 0.6292029619216919, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.9375, |
| "step": 280 |
| }, |
| { |
| "completion_length": 31.390625, |
| "epoch": 0.31118493909191586, |
| "grad_norm": 0.4693773090839386, |
| "kl": 0.02471923828125, |
| "learning_rate": 8.243749999999999e-07, |
| "loss": 0.006131879985332489, |
| "reward": 2.087433636188507, |
| "reward_std": 0.34090328216552734, |
| "rewards/GDino": 0.7668008506298065, |
| "rewards/GIT": 0.41047535091638565, |
| "rewards/HPSv2": 0.28122520446777344, |
| "rewards/ORM": 0.6289321780204773, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -20.875, |
| "step": 281 |
| }, |
| { |
| "completion_length": 48.78125, |
| "epoch": 0.3122923588039867, |
| "grad_norm": 1.8625261783599854, |
| "kl": 0.02197265625, |
| "learning_rate": 8.2375e-07, |
| "loss": 0.011913509108126163, |
| "reward": 2.104038119316101, |
| "reward_std": 0.27633778750896454, |
| "rewards/GDino": 0.7975816428661346, |
| "rewards/GIT": 0.6180287599563599, |
| "rewards/HPSv2": 0.26529884338378906, |
| "rewards/ORM": 0.42312875390052795, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -20.875, |
| "step": 282 |
| }, |
| { |
| "completion_length": 37.53125, |
| "epoch": 0.3133997785160576, |
| "grad_norm": 0.5229981541633606, |
| "kl": 0.014862060546875, |
| "learning_rate": 8.23125e-07, |
| "loss": 0.020383010618388653, |
| "reward": 2.0289142727851868, |
| "reward_std": 0.30830617249011993, |
| "rewards/GDino": 0.7265625298023224, |
| "rewards/GIT": 0.4469078704714775, |
| "rewards/HPSv2": 0.24579811096191406, |
| "rewards/ORM": 0.6096457839012146, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.25, |
| "step": 283 |
| }, |
| { |
| "completion_length": 39.609375, |
| "epoch": 0.31450719822812845, |
| "grad_norm": 0.4969702661037445, |
| "kl": 0.02374267578125, |
| "learning_rate": 8.225e-07, |
| "loss": -0.000512864557094872, |
| "reward": 1.7745079398155212, |
| "reward_std": 0.2966308742761612, |
| "rewards/GDino": 0.717956930398941, |
| "rewards/GIT": 0.34320028126239777, |
| "rewards/HPSv2": 0.2720203399658203, |
| "rewards/ORM": 0.4413303881883621, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.8125, |
| "step": 284 |
| }, |
| { |
| "completion_length": 44.375, |
| "epoch": 0.31561461794019935, |
| "grad_norm": 0.7119840383529663, |
| "kl": 0.0361328125, |
| "learning_rate": 8.21875e-07, |
| "loss": 0.01481186505407095, |
| "reward": 1.5489041805267334, |
| "reward_std": 0.393330454826355, |
| "rewards/GDino": 0.6588811278343201, |
| "rewards/GIT": 0.18679189682006836, |
| "rewards/HPSv2": 0.2698822021484375, |
| "rewards/ORM": 0.4333488643169403, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.5625, |
| "step": 285 |
| }, |
| { |
| "completion_length": 37.34375, |
| "epoch": 0.3167220376522702, |
| "grad_norm": 0.5771685242652893, |
| "kl": 0.01953125, |
| "learning_rate": 8.2125e-07, |
| "loss": -0.01632743887603283, |
| "reward": 2.1853543519973755, |
| "reward_std": 0.4599858373403549, |
| "rewards/GDino": 0.7937499582767487, |
| "rewards/GIT": 0.4548158571124077, |
| "rewards/HPSv2": 0.27312660217285156, |
| "rewards/ORM": 0.6636618673801422, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.5, |
| "step": 286 |
| }, |
| { |
| "completion_length": 32.453125, |
| "epoch": 0.3178294573643411, |
| "grad_norm": 0.4807415306568146, |
| "kl": 0.0235595703125, |
| "learning_rate": 8.20625e-07, |
| "loss": 0.009378439281135798, |
| "reward": 2.3349956274032593, |
| "reward_std": 0.3700142502784729, |
| "rewards/GDino": 0.7700349688529968, |
| "rewards/GIT": 0.5914836376905441, |
| "rewards/HPSv2": 0.2726278305053711, |
| "rewards/ORM": 0.7008491158485413, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.375, |
| "step": 287 |
| }, |
| { |
| "completion_length": 35.734375, |
| "epoch": 0.31893687707641194, |
| "grad_norm": 0.5664615035057068, |
| "kl": 0.0255126953125, |
| "learning_rate": 8.199999999999999e-07, |
| "loss": 0.0027993861585855484, |
| "reward": 2.357239305973053, |
| "reward_std": 0.34706495702266693, |
| "rewards/GDino": 0.7578125, |
| "rewards/GIT": 0.4684627875685692, |
| "rewards/HPSv2": 0.25521278381347656, |
| "rewards/ORM": 0.8757513463497162, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.625, |
| "step": 288 |
| }, |
| { |
| "completion_length": 39.1875, |
| "epoch": 0.32004429678848284, |
| "grad_norm": 0.8079530000686646, |
| "kl": 0.01953125, |
| "learning_rate": 8.193749999999999e-07, |
| "loss": -0.018345186486840248, |
| "reward": 2.2459497451782227, |
| "reward_std": 0.36141130328178406, |
| "rewards/GDino": 0.7167181968688965, |
| "rewards/GIT": 0.463926300406456, |
| "rewards/HPSv2": 0.2644481658935547, |
| "rewards/ORM": 0.8008571267127991, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -20.875, |
| "step": 289 |
| }, |
| { |
| "completion_length": 49.59375, |
| "epoch": 0.3211517165005537, |
| "grad_norm": 0.6423276662826538, |
| "kl": 0.02008056640625, |
| "learning_rate": 8.187499999999999e-07, |
| "loss": 0.004360657767392695, |
| "reward": 2.068402886390686, |
| "reward_std": 0.3118213042616844, |
| "rewards/GDino": 0.6897880434989929, |
| "rewards/GIT": 0.3221408724784851, |
| "rewards/HPSv2": 0.2720451354980469, |
| "rewards/ORM": 0.7844289541244507, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.3125, |
| "step": 290 |
| }, |
| { |
| "completion_length": 40.578125, |
| "epoch": 0.3222591362126246, |
| "grad_norm": 1.5936462879180908, |
| "kl": 0.02667236328125, |
| "learning_rate": 8.18125e-07, |
| "loss": 0.00672217458486557, |
| "reward": 1.89811110496521, |
| "reward_std": 0.3131570816040039, |
| "rewards/GDino": 0.7552083432674408, |
| "rewards/GIT": 0.38471363484859467, |
| "rewards/HPSv2": 0.26104736328125, |
| "rewards/ORM": 0.4971417486667633, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.0625, |
| "step": 291 |
| }, |
| { |
| "completion_length": 52.265625, |
| "epoch": 0.3233665559246955, |
| "grad_norm": 0.5862070918083191, |
| "kl": 0.0191650390625, |
| "learning_rate": 8.175e-07, |
| "loss": 0.0030853250063955784, |
| "reward": 2.2179116010665894, |
| "reward_std": 0.354862704873085, |
| "rewards/GDino": 0.8304687440395355, |
| "rewards/GIT": 0.5982599407434464, |
| "rewards/HPSv2": 0.26502227783203125, |
| "rewards/ORM": 0.5241607576608658, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.0, |
| "step": 292 |
| }, |
| { |
| "completion_length": 44.359375, |
| "epoch": 0.32447397563676633, |
| "grad_norm": 0.49480700492858887, |
| "kl": 0.025146484375, |
| "learning_rate": 8.16875e-07, |
| "loss": 0.00254095159471035, |
| "reward": 2.3431068062782288, |
| "reward_std": 0.2980917990207672, |
| "rewards/GDino": 0.7105664908885956, |
| "rewards/GIT": 0.45115914195775986, |
| "rewards/HPSv2": 0.2673187255859375, |
| "rewards/ORM": 0.9140625, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -20.9375, |
| "step": 293 |
| }, |
| { |
| "completion_length": 37.453125, |
| "epoch": 0.32558139534883723, |
| "grad_norm": 0.567101776599884, |
| "kl": 0.031494140625, |
| "learning_rate": 8.1625e-07, |
| "loss": -0.0005339896306395531, |
| "reward": 2.6543023586273193, |
| "reward_std": 0.3742424249649048, |
| "rewards/GDino": 0.8531249761581421, |
| "rewards/GIT": 0.5934174060821533, |
| "rewards/HPSv2": 0.2702598571777344, |
| "rewards/ORM": 0.9375, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.625, |
| "step": 294 |
| }, |
| { |
| "completion_length": 42.46875, |
| "epoch": 0.3266888150609081, |
| "grad_norm": 0.7111007571220398, |
| "kl": 0.0341796875, |
| "learning_rate": 8.15625e-07, |
| "loss": 0.006241308408789337, |
| "reward": 1.6757753491401672, |
| "reward_std": 0.4623751789331436, |
| "rewards/GDino": 0.6648381352424622, |
| "rewards/GIT": 0.07737746834754944, |
| "rewards/HPSv2": 0.2832469940185547, |
| "rewards/ORM": 0.6503127217292786, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.75, |
| "step": 295 |
| }, |
| { |
| "completion_length": 36.21875, |
| "epoch": 0.327796234772979, |
| "grad_norm": 0.563860297203064, |
| "kl": 0.02325439453125, |
| "learning_rate": 8.149999999999999e-07, |
| "loss": 0.019053890835493803, |
| "reward": 2.774321436882019, |
| "reward_std": 0.19439150393009186, |
| "rewards/GDino": 0.971875011920929, |
| "rewards/GIT": 0.7940655946731567, |
| "rewards/HPSv2": 0.2748088836669922, |
| "rewards/ORM": 0.733571857213974, |
| "self_certainty_semantic": -25.625, |
| "self_certainty_token": -20.3125, |
| "step": 296 |
| }, |
| { |
| "completion_length": 35.34375, |
| "epoch": 0.3289036544850498, |
| "grad_norm": 0.5320677161216736, |
| "kl": 0.02813720703125, |
| "learning_rate": 8.143749999999999e-07, |
| "loss": -8.378783240914345e-05, |
| "reward": 1.9089605808258057, |
| "reward_std": 0.33931228518486023, |
| "rewards/GDino": 0.7917934358119965, |
| "rewards/GIT": 0.2872927859425545, |
| "rewards/HPSv2": 0.26018333435058594, |
| "rewards/ORM": 0.5696910321712494, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.125, |
| "step": 297 |
| }, |
| { |
| "completion_length": 39.84375, |
| "epoch": 0.3300110741971207, |
| "grad_norm": 0.6157541871070862, |
| "kl": 0.02362060546875, |
| "learning_rate": 8.137499999999999e-07, |
| "loss": 0.0025934623554348946, |
| "reward": 2.2229180932044983, |
| "reward_std": 0.28714166209101677, |
| "rewards/GDino": 0.8539158701896667, |
| "rewards/GIT": 0.6188642829656601, |
| "rewards/HPSv2": 0.2704963684082031, |
| "rewards/ORM": 0.47964154183864594, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.4375, |
| "step": 298 |
| }, |
| { |
| "completion_length": 40.296875, |
| "epoch": 0.33111849390919157, |
| "grad_norm": 0.5113131403923035, |
| "kl": 0.016204833984375, |
| "learning_rate": 8.131249999999999e-07, |
| "loss": -0.00016327621415257454, |
| "reward": 2.4606435298919678, |
| "reward_std": 0.4053623676300049, |
| "rewards/GDino": 0.8729166686534882, |
| "rewards/GIT": 0.7344014346599579, |
| "rewards/HPSv2": 0.2575702667236328, |
| "rewards/ORM": 0.5957551002502441, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.3125, |
| "step": 299 |
| }, |
| { |
| "completion_length": 34.75, |
| "epoch": 0.33222591362126247, |
| "grad_norm": 0.7174614071846008, |
| "kl": 0.02789306640625, |
| "learning_rate": 8.125e-07, |
| "loss": -0.008093067444860935, |
| "reward": 2.5811712741851807, |
| "reward_std": 0.26995332539081573, |
| "rewards/GDino": 0.8097889721393585, |
| "rewards/GIT": 0.7062003910541534, |
| "rewards/HPSv2": 0.2668323516845703, |
| "rewards/ORM": 0.7983495891094208, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.25, |
| "step": 300 |
| }, |
| { |
| "completion_length": 40.921875, |
| "epoch": 0.3333333333333333, |
| "grad_norm": 0.5175552368164062, |
| "kl": 0.021728515625, |
| "learning_rate": 8.11875e-07, |
| "loss": -0.0006439080461859703, |
| "reward": 2.1311718225479126, |
| "reward_std": 0.41609956324100494, |
| "rewards/GDino": 0.7800666689872742, |
| "rewards/GIT": 0.42304590344429016, |
| "rewards/HPSv2": 0.24529266357421875, |
| "rewards/ORM": 0.6827665567398071, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.75, |
| "step": 301 |
| }, |
| { |
| "completion_length": 39.75, |
| "epoch": 0.3344407530454042, |
| "grad_norm": 0.4989144802093506, |
| "kl": 0.02520751953125, |
| "learning_rate": 8.1125e-07, |
| "loss": -0.0295205507427454, |
| "reward": 2.2222087383270264, |
| "reward_std": 0.28523801267147064, |
| "rewards/GDino": 0.807892918586731, |
| "rewards/GIT": 0.4448024183511734, |
| "rewards/HPSv2": 0.2647838592529297, |
| "rewards/ORM": 0.704729437828064, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.5, |
| "step": 302 |
| }, |
| { |
| "completion_length": 34.4375, |
| "epoch": 0.33554817275747506, |
| "grad_norm": 2.3144850730895996, |
| "kl": 0.0274658203125, |
| "learning_rate": 8.10625e-07, |
| "loss": 3.9184000343084335e-05, |
| "reward": 1.9459970593452454, |
| "reward_std": 0.3054793253540993, |
| "rewards/GDino": 0.6584998369216919, |
| "rewards/GIT": 0.17534568905830383, |
| "rewards/HPSv2": 0.2799644470214844, |
| "rewards/ORM": 0.8321870565414429, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.0, |
| "step": 303 |
| }, |
| { |
| "completion_length": 35.15625, |
| "epoch": 0.33665559246954596, |
| "grad_norm": 0.49459758400917053, |
| "kl": 0.01934814453125, |
| "learning_rate": 8.1e-07, |
| "loss": 0.024405019357800484, |
| "reward": 2.1167579889297485, |
| "reward_std": 0.40585383772850037, |
| "rewards/GDino": 0.729687511920929, |
| "rewards/GIT": 0.4396758899092674, |
| "rewards/HPSv2": 0.27677345275878906, |
| "rewards/ORM": 0.670621246099472, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.6875, |
| "step": 304 |
| }, |
| { |
| "completion_length": 43.234375, |
| "epoch": 0.3377630121816168, |
| "grad_norm": 0.4376680552959442, |
| "kl": 0.02001953125, |
| "learning_rate": 8.093749999999999e-07, |
| "loss": 0.016018358757719398, |
| "reward": 1.974160075187683, |
| "reward_std": 0.36807504296302795, |
| "rewards/GDino": 0.7174843549728394, |
| "rewards/GIT": 0.35652345418930054, |
| "rewards/HPSv2": 0.25952720642089844, |
| "rewards/ORM": 0.640625, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.8125, |
| "step": 305 |
| }, |
| { |
| "completion_length": 41.0, |
| "epoch": 0.3388704318936877, |
| "grad_norm": 0.4617772400379181, |
| "kl": 0.02471923828125, |
| "learning_rate": 8.087499999999999e-07, |
| "loss": -0.016240317840129137, |
| "reward": 1.7467725276947021, |
| "reward_std": 0.5162160992622375, |
| "rewards/GDino": 0.6519080102443695, |
| "rewards/GIT": 0.19782309979200363, |
| "rewards/HPSv2": 0.27204132080078125, |
| "rewards/ORM": 0.625, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.625, |
| "step": 306 |
| }, |
| { |
| "completion_length": 43.171875, |
| "epoch": 0.3399778516057586, |
| "grad_norm": 0.5335010886192322, |
| "kl": 0.02056884765625, |
| "learning_rate": 8.08125e-07, |
| "loss": 0.01701421057805419, |
| "reward": 2.1066606044769287, |
| "reward_std": 0.22635112702846527, |
| "rewards/GDino": 0.7486607134342194, |
| "rewards/GIT": 0.6341286897659302, |
| "rewards/HPSv2": 0.25512123107910156, |
| "rewards/ORM": 0.46875, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.6875, |
| "step": 307 |
| }, |
| { |
| "completion_length": 40.453125, |
| "epoch": 0.34108527131782945, |
| "grad_norm": 0.524975061416626, |
| "kl": 0.02642822265625, |
| "learning_rate": 8.075e-07, |
| "loss": -0.009714136831462383, |
| "reward": 2.396199941635132, |
| "reward_std": 0.33491872251033783, |
| "rewards/GDino": 0.9330095648765564, |
| "rewards/GIT": 0.6121770441532135, |
| "rewards/HPSv2": 0.2553081512451172, |
| "rewards/ORM": 0.5957051813602448, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.75, |
| "step": 308 |
| }, |
| { |
| "completion_length": 35.671875, |
| "epoch": 0.34219269102990035, |
| "grad_norm": 0.6639419198036194, |
| "kl": 0.0311279296875, |
| "learning_rate": 8.06875e-07, |
| "loss": -0.029759632423520088, |
| "reward": 1.9770995378494263, |
| "reward_std": 0.2913192883133888, |
| "rewards/GDino": 0.7039497792720795, |
| "rewards/GIT": 0.32885563373565674, |
| "rewards/HPSv2": 0.2879142761230469, |
| "rewards/ORM": 0.656379908323288, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.375, |
| "step": 309 |
| }, |
| { |
| "completion_length": 34.53125, |
| "epoch": 0.3433001107419712, |
| "grad_norm": 0.5760204195976257, |
| "kl": 0.02392578125, |
| "learning_rate": 8.0625e-07, |
| "loss": -0.006422802805900574, |
| "reward": 2.6306651830673218, |
| "reward_std": 0.45363010466098785, |
| "rewards/GDino": 0.8823047280311584, |
| "rewards/GIT": 0.6377567946910858, |
| "rewards/HPSv2": 0.25002479553222656, |
| "rewards/ORM": 0.8605788052082062, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.375, |
| "step": 310 |
| }, |
| { |
| "completion_length": 37.0, |
| "epoch": 0.3444075304540421, |
| "grad_norm": 0.46515122056007385, |
| "kl": 0.0203857421875, |
| "learning_rate": 8.05625e-07, |
| "loss": 0.004794539883732796, |
| "reward": 2.3176556825637817, |
| "reward_std": 0.28090114146471024, |
| "rewards/GDino": 0.7750000357627869, |
| "rewards/GIT": 0.5293775945901871, |
| "rewards/HPSv2": 0.25220680236816406, |
| "rewards/ORM": 0.7610713541507721, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -20.9375, |
| "step": 311 |
| }, |
| { |
| "completion_length": 39.09375, |
| "epoch": 0.34551495016611294, |
| "grad_norm": 1.104447841644287, |
| "kl": 0.02801513671875, |
| "learning_rate": 8.05e-07, |
| "loss": 0.007791162468492985, |
| "reward": 2.1565089225769043, |
| "reward_std": 0.5575137436389923, |
| "rewards/GDino": 0.8127583563327789, |
| "rewards/GIT": 0.4446650370955467, |
| "rewards/HPSv2": 0.26712608337402344, |
| "rewards/ORM": 0.6319593787193298, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.5625, |
| "step": 312 |
| }, |
| { |
| "completion_length": 45.46875, |
| "epoch": 0.34662236987818384, |
| "grad_norm": 0.5203121304512024, |
| "kl": 0.072509765625, |
| "learning_rate": 8.043749999999999e-07, |
| "loss": 0.012506457045674324, |
| "reward": 1.833688497543335, |
| "reward_std": 0.3239995241165161, |
| "rewards/GDino": 0.6667128503322601, |
| "rewards/GIT": 0.2989926040172577, |
| "rewards/HPSv2": 0.2667713165283203, |
| "rewards/ORM": 0.6012117862701416, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.1875, |
| "step": 313 |
| }, |
| { |
| "completion_length": 40.03125, |
| "epoch": 0.3477297895902547, |
| "grad_norm": 0.5220968127250671, |
| "kl": 0.029296875, |
| "learning_rate": 8.037499999999999e-07, |
| "loss": 0.015390763524919748, |
| "reward": 1.8818817138671875, |
| "reward_std": 0.3023659586906433, |
| "rewards/GDino": 0.7861979305744171, |
| "rewards/GIT": 0.461479514837265, |
| "rewards/HPSv2": 0.24705886840820312, |
| "rewards/ORM": 0.3871453106403351, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.0, |
| "step": 314 |
| }, |
| { |
| "completion_length": 40.84375, |
| "epoch": 0.3488372093023256, |
| "grad_norm": 0.6556162238121033, |
| "kl": 0.02447509765625, |
| "learning_rate": 8.031249999999999e-07, |
| "loss": -0.011867262073792517, |
| "reward": 2.328538656234741, |
| "reward_std": 0.3933458775281906, |
| "rewards/GDino": 0.8490088582038879, |
| "rewards/GIT": 0.5727739632129669, |
| "rewards/HPSv2": 0.27246856689453125, |
| "rewards/ORM": 0.6342871189117432, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.6875, |
| "step": 315 |
| }, |
| { |
| "completion_length": 45.859375, |
| "epoch": 0.34994462901439644, |
| "grad_norm": 0.9298399686813354, |
| "kl": 0.0322265625, |
| "learning_rate": 8.024999999999999e-07, |
| "loss": 0.004026470240205526, |
| "reward": 2.328371286392212, |
| "reward_std": 0.3506554663181305, |
| "rewards/GDino": 0.6822916865348816, |
| "rewards/GIT": 0.5789465606212616, |
| "rewards/HPSv2": 0.25463294982910156, |
| "rewards/ORM": 0.8125, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.375, |
| "step": 316 |
| }, |
| { |
| "completion_length": 41.609375, |
| "epoch": 0.35105204872646734, |
| "grad_norm": 1.1766077280044556, |
| "kl": 0.03387451171875, |
| "learning_rate": 8.018749999999999e-07, |
| "loss": 0.0020321097690612078, |
| "reward": 1.7646815776824951, |
| "reward_std": 0.3271322548389435, |
| "rewards/GDino": 0.7106184661388397, |
| "rewards/GIT": 0.21883262693881989, |
| "rewards/HPSv2": 0.28487586975097656, |
| "rewards/ORM": 0.5503546893596649, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.1875, |
| "step": 317 |
| }, |
| { |
| "completion_length": 36.171875, |
| "epoch": 0.3521594684385382, |
| "grad_norm": 0.5149301290512085, |
| "kl": 0.025634765625, |
| "learning_rate": 8.0125e-07, |
| "loss": -0.03277874179184437, |
| "reward": 2.138643741607666, |
| "reward_std": 0.3318810313940048, |
| "rewards/GDino": 0.7371024787425995, |
| "rewards/GIT": 0.4322855994105339, |
| "rewards/HPSv2": 0.26688194274902344, |
| "rewards/ORM": 0.7023736536502838, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.5, |
| "step": 318 |
| }, |
| { |
| "completion_length": 37.6875, |
| "epoch": 0.3532668881506091, |
| "grad_norm": 1.4863990545272827, |
| "kl": 0.02178955078125, |
| "learning_rate": 8.00625e-07, |
| "loss": -0.002279828302562237, |
| "reward": 1.8946493268013, |
| "reward_std": 0.27233169972896576, |
| "rewards/GDino": 0.7110094726085663, |
| "rewards/GIT": 0.23650505393743515, |
| "rewards/HPSv2": 0.2566356658935547, |
| "rewards/ORM": 0.6904990971088409, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.125, |
| "step": 319 |
| }, |
| { |
| "completion_length": 47.5, |
| "epoch": 0.35437430786268, |
| "grad_norm": 0.43619126081466675, |
| "kl": 0.023193359375, |
| "learning_rate": 8e-07, |
| "loss": 0.003124894807115197, |
| "reward": 2.0905760526657104, |
| "reward_std": 0.3077670633792877, |
| "rewards/GDino": 0.7233002185821533, |
| "rewards/GIT": 0.5371533036231995, |
| "rewards/HPSv2": 0.2641429901123047, |
| "rewards/ORM": 0.5659796893596649, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.625, |
| "step": 320 |
| }, |
| { |
| "completion_length": 45.390625, |
| "epoch": 0.3554817275747508, |
| "grad_norm": 0.5557716488838196, |
| "kl": 0.037109375, |
| "learning_rate": 7.993749999999999e-07, |
| "loss": -0.010770568624138832, |
| "reward": 2.225831151008606, |
| "reward_std": 0.3986949622631073, |
| "rewards/GDino": 0.8189918994903564, |
| "rewards/GIT": 0.45973852276802063, |
| "rewards/HPSv2": 0.2627296447753906, |
| "rewards/ORM": 0.6843710243701935, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -20.875, |
| "step": 321 |
| }, |
| { |
| "completion_length": 29.953125, |
| "epoch": 0.35658914728682173, |
| "grad_norm": 0.6458404660224915, |
| "kl": 0.03125, |
| "learning_rate": 7.9875e-07, |
| "loss": -0.01060107909142971, |
| "reward": 2.633362889289856, |
| "reward_std": 0.4656294733285904, |
| "rewards/GDino": 0.9093749523162842, |
| "rewards/GIT": 0.6372530907392502, |
| "rewards/HPSv2": 0.2742347717285156, |
| "rewards/ORM": 0.8124999701976776, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.125, |
| "step": 322 |
| }, |
| { |
| "completion_length": 33.609375, |
| "epoch": 0.3576965669988926, |
| "grad_norm": 0.7263385653495789, |
| "kl": 0.03936767578125, |
| "learning_rate": 7.98125e-07, |
| "loss": 0.017037604935467243, |
| "reward": 2.223346173763275, |
| "reward_std": 0.27071254700422287, |
| "rewards/GDino": 0.8520833551883698, |
| "rewards/GIT": 0.49767685681581497, |
| "rewards/HPSv2": 0.26546478271484375, |
| "rewards/ORM": 0.6081212162971497, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.8125, |
| "step": 323 |
| }, |
| { |
| "completion_length": 44.796875, |
| "epoch": 0.3588039867109635, |
| "grad_norm": 0.6808414459228516, |
| "kl": 0.0328369140625, |
| "learning_rate": 7.975e-07, |
| "loss": 0.006063681095838547, |
| "reward": 2.1634711027145386, |
| "reward_std": 0.39905455708503723, |
| "rewards/GDino": 0.7807291746139526, |
| "rewards/GIT": 0.5616367310285568, |
| "rewards/HPSv2": 0.2618141174316406, |
| "rewards/ORM": 0.5592910945415497, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -20.5, |
| "step": 324 |
| }, |
| { |
| "completion_length": 34.78125, |
| "epoch": 0.3599114064230343, |
| "grad_norm": 0.6440313458442688, |
| "kl": 0.0216064453125, |
| "learning_rate": 7.96875e-07, |
| "loss": 0.0036031128838658333, |
| "reward": 2.2728426456451416, |
| "reward_std": 0.36220329999923706, |
| "rewards/GDino": 0.8558787703514099, |
| "rewards/GIT": 0.5083786845207214, |
| "rewards/HPSv2": 0.24144363403320312, |
| "rewards/ORM": 0.6671415269374847, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.75, |
| "step": 325 |
| }, |
| { |
| "completion_length": 40.875, |
| "epoch": 0.3610188261351052, |
| "grad_norm": 3325586688.0, |
| "kl": 7208960.011962891, |
| "learning_rate": 7.9625e-07, |
| "loss": 72052.51197006088, |
| "reward": 2.1175559759140015, |
| "reward_std": 0.23405619710683823, |
| "rewards/GDino": 0.7784741222858429, |
| "rewards/GIT": 0.4605119079351425, |
| "rewards/HPSv2": 0.2704486846923828, |
| "rewards/ORM": 0.6081212311983109, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.8125, |
| "step": 326 |
| }, |
| { |
| "completion_length": 34.125, |
| "epoch": 0.36212624584717606, |
| "grad_norm": 2.4330286979675293, |
| "kl": 0.0244140625, |
| "learning_rate": 7.95625e-07, |
| "loss": 0.006208475679159164, |
| "reward": 2.148754298686981, |
| "reward_std": 0.17882763594388962, |
| "rewards/GDino": 0.6925546824932098, |
| "rewards/GIT": 0.5509190559387207, |
| "rewards/HPSv2": 0.2755470275878906, |
| "rewards/ORM": 0.6297334432601929, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.5625, |
| "step": 327 |
| }, |
| { |
| "completion_length": 28.40625, |
| "epoch": 0.36323366555924697, |
| "grad_norm": 0.5411335229873657, |
| "kl": 0.04150390625, |
| "learning_rate": 7.95e-07, |
| "loss": 0.002802429720759392, |
| "reward": 2.179747223854065, |
| "reward_std": 0.22097515314817429, |
| "rewards/GDino": 0.796875, |
| "rewards/GIT": 0.4172445684671402, |
| "rewards/HPSv2": 0.29375267028808594, |
| "rewards/ORM": 0.671875, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.9375, |
| "step": 328 |
| }, |
| { |
| "completion_length": 49.09375, |
| "epoch": 0.3643410852713178, |
| "grad_norm": 0.6255419254302979, |
| "kl": 0.01898193359375, |
| "learning_rate": 7.94375e-07, |
| "loss": -0.01904204487800598, |
| "reward": 2.2398595809936523, |
| "reward_std": 0.2658727616071701, |
| "rewards/GDino": 0.7032954692840576, |
| "rewards/GIT": 0.4768924117088318, |
| "rewards/HPSv2": 0.2772178649902344, |
| "rewards/ORM": 0.7824538052082062, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.625, |
| "step": 329 |
| }, |
| { |
| "completion_length": 40.21875, |
| "epoch": 0.3654485049833887, |
| "grad_norm": 0.5795601606369019, |
| "kl": 0.03448486328125, |
| "learning_rate": 7.937499999999999e-07, |
| "loss": -0.028916708659380674, |
| "reward": 1.9032581448554993, |
| "reward_std": 0.3265683054924011, |
| "rewards/GDino": 0.7300936579704285, |
| "rewards/GIT": 0.3637319356203079, |
| "rewards/HPSv2": 0.2578239440917969, |
| "rewards/ORM": 0.5516084283590317, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.125, |
| "step": 330 |
| }, |
| { |
| "completion_length": 39.0625, |
| "epoch": 0.36655592469545956, |
| "grad_norm": 0.5067495703697205, |
| "kl": 0.02813720703125, |
| "learning_rate": 7.931249999999999e-07, |
| "loss": -0.009640714153647423, |
| "reward": 2.4922925233840942, |
| "reward_std": 0.3301289528608322, |
| "rewards/GDino": 0.870312511920929, |
| "rewards/GIT": 0.7084252834320068, |
| "rewards/HPSv2": 0.2510967254638672, |
| "rewards/ORM": 0.6624580323696136, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.625, |
| "step": 331 |
| }, |
| { |
| "completion_length": 47.796875, |
| "epoch": 0.36766334440753046, |
| "grad_norm": 0.671696662902832, |
| "kl": 0.02166748046875, |
| "learning_rate": 7.924999999999999e-07, |
| "loss": -0.004493948072195053, |
| "reward": 1.7546527981758118, |
| "reward_std": 0.38349293172359467, |
| "rewards/GDino": 0.7021358013153076, |
| "rewards/GIT": 0.32257433235645294, |
| "rewards/HPSv2": 0.25150489807128906, |
| "rewards/ORM": 0.47843775153160095, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.0625, |
| "step": 332 |
| }, |
| { |
| "completion_length": 45.03125, |
| "epoch": 0.3687707641196013, |
| "grad_norm": 0.5712724924087524, |
| "kl": 0.0302734375, |
| "learning_rate": 7.918749999999999e-07, |
| "loss": 0.019796861335635185, |
| "reward": 2.2836796045303345, |
| "reward_std": 0.3291999250650406, |
| "rewards/GDino": 0.7703441679477692, |
| "rewards/GIT": 0.4380648583173752, |
| "rewards/HPSv2": 0.2811241149902344, |
| "rewards/ORM": 0.7941466867923737, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.5625, |
| "step": 333 |
| }, |
| { |
| "completion_length": 34.546875, |
| "epoch": 0.3698781838316722, |
| "grad_norm": 0.4348975718021393, |
| "kl": 0.01947021484375, |
| "learning_rate": 7.912499999999999e-07, |
| "loss": 0.032695122761651874, |
| "reward": 2.472940444946289, |
| "reward_std": 0.2854895293712616, |
| "rewards/GDino": 0.8704217672348022, |
| "rewards/GIT": 0.5610889345407486, |
| "rewards/HPSv2": 0.25870513916015625, |
| "rewards/ORM": 0.7827245891094208, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.125, |
| "step": 334 |
| }, |
| { |
| "completion_length": 45.109375, |
| "epoch": 0.3709856035437431, |
| "grad_norm": 0.5861725807189941, |
| "kl": 0.03179931640625, |
| "learning_rate": 7.90625e-07, |
| "loss": 0.001954459585249424, |
| "reward": 2.100743532180786, |
| "reward_std": 0.34255510568618774, |
| "rewards/GDino": 0.8146567642688751, |
| "rewards/GIT": 0.3769591152667999, |
| "rewards/HPSv2": 0.27979469299316406, |
| "rewards/ORM": 0.6293327808380127, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.625, |
| "step": 335 |
| }, |
| { |
| "completion_length": 47.125, |
| "epoch": 0.37209302325581395, |
| "grad_norm": 0.47423750162124634, |
| "kl": 0.02886962890625, |
| "learning_rate": 7.9e-07, |
| "loss": -0.0015065963380038738, |
| "reward": 2.343234896659851, |
| "reward_std": 0.2731991782784462, |
| "rewards/GDino": 0.8291987180709839, |
| "rewards/GIT": 0.7442143559455872, |
| "rewards/HPSv2": 0.27597999572753906, |
| "rewards/ORM": 0.49384191632270813, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.6875, |
| "step": 336 |
| }, |
| { |
| "completion_length": 47.265625, |
| "epoch": 0.37320044296788485, |
| "grad_norm": 0.45597055554389954, |
| "kl": 0.02423095703125, |
| "learning_rate": 7.893750000000001e-07, |
| "loss": 0.025022207759320736, |
| "reward": 1.9880129098892212, |
| "reward_std": 0.31796349585056305, |
| "rewards/GDino": 0.7098958492279053, |
| "rewards/GIT": 0.37319713830947876, |
| "rewards/HPSv2": 0.2521495819091797, |
| "rewards/ORM": 0.6527703106403351, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.3125, |
| "step": 337 |
| }, |
| { |
| "completion_length": 37.234375, |
| "epoch": 0.3743078626799557, |
| "grad_norm": 0.5559476017951965, |
| "kl": 0.03076171875, |
| "learning_rate": 7.8875e-07, |
| "loss": -0.0021500587463378906, |
| "reward": 1.8510813117027283, |
| "reward_std": 0.4456440210342407, |
| "rewards/GDino": 0.7363041043281555, |
| "rewards/GIT": 0.4989383816719055, |
| "rewards/HPSv2": 0.2646770477294922, |
| "rewards/ORM": 0.3511618450284004, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.4375, |
| "step": 338 |
| }, |
| { |
| "completion_length": 37.109375, |
| "epoch": 0.3754152823920266, |
| "grad_norm": 0.7247738838195801, |
| "kl": 0.025390625, |
| "learning_rate": 7.88125e-07, |
| "loss": 0.005743648856878281, |
| "reward": 2.28269100189209, |
| "reward_std": 0.4758919030427933, |
| "rewards/GDino": 0.7302083075046539, |
| "rewards/GIT": 0.5845783352851868, |
| "rewards/HPSv2": 0.26825904846191406, |
| "rewards/ORM": 0.6996453106403351, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.625, |
| "step": 339 |
| }, |
| { |
| "completion_length": 33.828125, |
| "epoch": 0.37652270210409744, |
| "grad_norm": 0.46468424797058105, |
| "kl": 0.028076171875, |
| "learning_rate": 7.875e-07, |
| "loss": 0.0004999474622309208, |
| "reward": 2.349074602127075, |
| "reward_std": 0.37045538425445557, |
| "rewards/GDino": 0.7835937142372131, |
| "rewards/GIT": 0.4427451193332672, |
| "rewards/HPSv2": 0.2726478576660156, |
| "rewards/ORM": 0.8500878810882568, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.375, |
| "step": 340 |
| }, |
| { |
| "completion_length": 37.140625, |
| "epoch": 0.37763012181616834, |
| "grad_norm": 0.978127121925354, |
| "kl": 0.033447265625, |
| "learning_rate": 7.86875e-07, |
| "loss": -0.014791712863370776, |
| "reward": 1.7075093984603882, |
| "reward_std": 0.4154247045516968, |
| "rewards/GDino": 0.6365624964237213, |
| "rewards/GIT": 0.23427681624889374, |
| "rewards/HPSv2": 0.2517547607421875, |
| "rewards/ORM": 0.5849153995513916, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.5, |
| "step": 341 |
| }, |
| { |
| "completion_length": 39.640625, |
| "epoch": 0.3787375415282392, |
| "grad_norm": 0.5854309797286987, |
| "kl": 0.02081298828125, |
| "learning_rate": 7.8625e-07, |
| "loss": 0.008088980801403522, |
| "reward": 1.816591739654541, |
| "reward_std": 0.43020693957805634, |
| "rewards/GDino": 0.6936458945274353, |
| "rewards/GIT": 0.40889062732458115, |
| "rewards/HPSv2": 0.2503395080566406, |
| "rewards/ORM": 0.4637157917022705, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.3125, |
| "step": 342 |
| }, |
| { |
| "completion_length": 33.859375, |
| "epoch": 0.3798449612403101, |
| "grad_norm": 0.5576260089874268, |
| "kl": 0.02655029296875, |
| "learning_rate": 7.85625e-07, |
| "loss": 0.007988112978637218, |
| "reward": 2.101312041282654, |
| "reward_std": 0.2812753766775131, |
| "rewards/GDino": 0.7948440611362457, |
| "rewards/GIT": 0.2880048602819443, |
| "rewards/HPSv2": 0.2610511779785156, |
| "rewards/ORM": 0.7574118673801422, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.4375, |
| "step": 343 |
| }, |
| { |
| "completion_length": 39.25, |
| "epoch": 0.38095238095238093, |
| "grad_norm": 0.546230673789978, |
| "kl": 0.0289306640625, |
| "learning_rate": 7.85e-07, |
| "loss": 0.010598460678011179, |
| "reward": 2.3708373308181763, |
| "reward_std": 0.32361893355846405, |
| "rewards/GDino": 0.7421875298023224, |
| "rewards/GIT": 0.7563790380954742, |
| "rewards/HPSv2": 0.2398090362548828, |
| "rewards/ORM": 0.632461816072464, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.625, |
| "step": 344 |
| }, |
| { |
| "completion_length": 43.375, |
| "epoch": 0.38205980066445183, |
| "grad_norm": 0.4513307213783264, |
| "kl": 0.0263671875, |
| "learning_rate": 7.84375e-07, |
| "loss": -0.004892995581030846, |
| "reward": 1.5629626512527466, |
| "reward_std": 0.5399304926395416, |
| "rewards/GDino": 0.6694765985012054, |
| "rewards/GIT": 0.28154174983501434, |
| "rewards/HPSv2": 0.23694419860839844, |
| "rewards/ORM": 0.375, |
| "self_certainty_semantic": -24.8125, |
| "self_certainty_token": -21.9375, |
| "step": 345 |
| }, |
| { |
| "completion_length": 40.375, |
| "epoch": 0.3831672203765227, |
| "grad_norm": 0.6939501762390137, |
| "kl": 0.0322265625, |
| "learning_rate": 7.837499999999999e-07, |
| "loss": -0.0008523159194737673, |
| "reward": 2.43539160490036, |
| "reward_std": 0.2919086888432503, |
| "rewards/GDino": 0.7796874940395355, |
| "rewards/GIT": 0.6580008417367935, |
| "rewards/HPSv2": 0.25712013244628906, |
| "rewards/ORM": 0.740583062171936, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.9375, |
| "step": 346 |
| }, |
| { |
| "completion_length": 40.0, |
| "epoch": 0.3842746400885936, |
| "grad_norm": 0.6701419353485107, |
| "kl": 0.03582763671875, |
| "learning_rate": 7.831249999999999e-07, |
| "loss": 0.002183706732466817, |
| "reward": 1.8965783715248108, |
| "reward_std": 0.35588257014751434, |
| "rewards/GDino": 0.6681874692440033, |
| "rewards/GIT": 0.30046548694372177, |
| "rewards/HPSv2": 0.27935791015625, |
| "rewards/ORM": 0.6485673785209656, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.375, |
| "step": 347 |
| }, |
| { |
| "completion_length": 53.6875, |
| "epoch": 0.3853820598006645, |
| "grad_norm": 0.5544401407241821, |
| "kl": 0.036376953125, |
| "learning_rate": 7.824999999999999e-07, |
| "loss": 0.017971528694033623, |
| "reward": 1.9810231924057007, |
| "reward_std": 0.416934609413147, |
| "rewards/GDino": 0.6915624737739563, |
| "rewards/GIT": 0.36784209311008453, |
| "rewards/HPSv2": 0.2577857971191406, |
| "rewards/ORM": 0.6638327836990356, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -20.625, |
| "step": 348 |
| }, |
| { |
| "completion_length": 41.96875, |
| "epoch": 0.3864894795127353, |
| "grad_norm": 0.9691958427429199, |
| "kl": 0.02850341796875, |
| "learning_rate": 7.818749999999999e-07, |
| "loss": 0.006191606633365154, |
| "reward": 1.978796124458313, |
| "reward_std": 0.2658599987626076, |
| "rewards/GDino": 0.7494639158248901, |
| "rewards/GIT": 0.3184107020497322, |
| "rewards/HPSv2": 0.2684211730957031, |
| "rewards/ORM": 0.6425002217292786, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.1875, |
| "step": 349 |
| }, |
| { |
| "completion_length": 42.1875, |
| "epoch": 0.3875968992248062, |
| "grad_norm": 5.835413455963135, |
| "kl": 0.04150390625, |
| "learning_rate": 7.812499999999999e-07, |
| "loss": -0.00010933470912277699, |
| "reward": 2.14810848236084, |
| "reward_std": 0.40148836374282837, |
| "rewards/GDino": 0.7505297660827637, |
| "rewards/GIT": 0.42371469736099243, |
| "rewards/HPSv2": 0.2747211456298828, |
| "rewards/ORM": 0.6991429328918457, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.125, |
| "step": 350 |
| }, |
| { |
| "completion_length": 35.765625, |
| "epoch": 0.38870431893687707, |
| "grad_norm": 0.4737742841243744, |
| "kl": 0.03106689453125, |
| "learning_rate": 7.806249999999999e-07, |
| "loss": 0.005023700650781393, |
| "reward": 2.1896114349365234, |
| "reward_std": 0.5353913605213165, |
| "rewards/GDino": 0.8089349865913391, |
| "rewards/GIT": 0.3830580413341522, |
| "rewards/HPSv2": 0.2619895935058594, |
| "rewards/ORM": 0.735628753900528, |
| "self_certainty_semantic": -24.75, |
| "self_certainty_token": -22.1875, |
| "step": 351 |
| }, |
| { |
| "completion_length": 34.25, |
| "epoch": 0.38981173864894797, |
| "grad_norm": 0.4819023907184601, |
| "kl": 0.0419921875, |
| "learning_rate": 7.799999999999999e-07, |
| "loss": 0.0019328959751874208, |
| "reward": 1.978476643562317, |
| "reward_std": 0.36969317495822906, |
| "rewards/GDino": 0.6918750107288361, |
| "rewards/GIT": 0.29961538314819336, |
| "rewards/HPSv2": 0.2604236602783203, |
| "rewards/ORM": 0.7265625, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.8125, |
| "step": 352 |
| }, |
| { |
| "completion_length": 45.15625, |
| "epoch": 0.3909191583610188, |
| "grad_norm": 0.5503346920013428, |
| "kl": 0.0274658203125, |
| "learning_rate": 7.793750000000001e-07, |
| "loss": 0.019882426131516695, |
| "reward": 1.7672075629234314, |
| "reward_std": 0.40149015188217163, |
| "rewards/GDino": 0.5861979126930237, |
| "rewards/GIT": 0.35189656913280487, |
| "rewards/HPSv2": 0.26621246337890625, |
| "rewards/ORM": 0.5629006028175354, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.0, |
| "step": 353 |
| }, |
| { |
| "completion_length": 30.78125, |
| "epoch": 0.3920265780730897, |
| "grad_norm": 0.47999808192253113, |
| "kl": 0.03118896484375, |
| "learning_rate": 7.787500000000001e-07, |
| "loss": 0.015973938745446503, |
| "reward": 2.0040203332901, |
| "reward_std": 0.5554044544696808, |
| "rewards/GDino": 0.666580930352211, |
| "rewards/GIT": 0.35180309414863586, |
| "rewards/HPSv2": 0.2794322967529297, |
| "rewards/ORM": 0.7062040567398071, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.1875, |
| "step": 354 |
| }, |
| { |
| "completion_length": 37.171875, |
| "epoch": 0.39313399778516056, |
| "grad_norm": 0.6516053080558777, |
| "kl": 0.03759765625, |
| "learning_rate": 7.78125e-07, |
| "loss": 0.02263426687568426, |
| "reward": 2.3876103162765503, |
| "reward_std": 0.2892715036869049, |
| "rewards/GDino": 0.7893574237823486, |
| "rewards/GIT": 0.6562651693820953, |
| "rewards/HPSv2": 0.2697620391845703, |
| "rewards/ORM": 0.6722257137298584, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.0625, |
| "step": 355 |
| }, |
| { |
| "completion_length": 47.40625, |
| "epoch": 0.39424141749723146, |
| "grad_norm": 0.5215231776237488, |
| "kl": 0.0260009765625, |
| "learning_rate": 7.775e-07, |
| "loss": -0.006568143609911203, |
| "reward": 2.1998695135116577, |
| "reward_std": 0.3379738926887512, |
| "rewards/GDino": 0.7524849772453308, |
| "rewards/GIT": 0.30117174983024597, |
| "rewards/HPSv2": 0.28067970275878906, |
| "rewards/ORM": 0.8655331134796143, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.5, |
| "step": 356 |
| }, |
| { |
| "completion_length": 39.484375, |
| "epoch": 0.3953488372093023, |
| "grad_norm": 0.44946345686912537, |
| "kl": 0.032470703125, |
| "learning_rate": 7.76875e-07, |
| "loss": -0.0038691945374011993, |
| "reward": 1.9695230722427368, |
| "reward_std": 0.3742541968822479, |
| "rewards/GDino": 0.6920569837093353, |
| "rewards/GIT": 0.31663452088832855, |
| "rewards/HPSv2": 0.2782859802246094, |
| "rewards/ORM": 0.6825457215309143, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.0625, |
| "step": 357 |
| }, |
| { |
| "completion_length": 44.4375, |
| "epoch": 0.3964562569213732, |
| "grad_norm": 0.5392388701438904, |
| "kl": 0.03759765625, |
| "learning_rate": 7.7625e-07, |
| "loss": -0.012778437230736017, |
| "reward": 2.211210250854492, |
| "reward_std": 0.4425947368144989, |
| "rewards/GDino": 0.7417187690734863, |
| "rewards/GIT": 0.5134639292955399, |
| "rewards/HPSv2": 0.26231956481933594, |
| "rewards/ORM": 0.6937080323696136, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.0, |
| "step": 358 |
| }, |
| { |
| "completion_length": 49.65625, |
| "epoch": 0.39756367663344405, |
| "grad_norm": 0.5134049654006958, |
| "kl": 0.02960205078125, |
| "learning_rate": 7.75625e-07, |
| "loss": -0.012967417016625404, |
| "reward": 1.412535309791565, |
| "reward_std": 0.45060451328754425, |
| "rewards/GDino": 0.6461889147758484, |
| "rewards/GIT": 0.19755424559116364, |
| "rewards/HPSv2": 0.2508354187011719, |
| "rewards/ORM": 0.31795668601989746, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.5625, |
| "step": 359 |
| }, |
| { |
| "completion_length": 35.609375, |
| "epoch": 0.39867109634551495, |
| "grad_norm": 1.2789876461029053, |
| "kl": 0.0343017578125, |
| "learning_rate": 7.75e-07, |
| "loss": -0.009992476087063551, |
| "reward": 2.073417007923126, |
| "reward_std": 0.38225309550762177, |
| "rewards/GDino": 0.7351497113704681, |
| "rewards/GIT": 0.5839340090751648, |
| "rewards/HPSv2": 0.27143287658691406, |
| "rewards/ORM": 0.48290038108825684, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.1875, |
| "step": 360 |
| }, |
| { |
| "completion_length": 39.578125, |
| "epoch": 0.3997785160575858, |
| "grad_norm": 0.5060411691665649, |
| "kl": 0.03515625, |
| "learning_rate": 7.74375e-07, |
| "loss": -0.005461947526782751, |
| "reward": 2.1116931438446045, |
| "reward_std": 0.4044407308101654, |
| "rewards/GDino": 0.7036125659942627, |
| "rewards/GIT": 0.44938327372074127, |
| "rewards/HPSv2": 0.2743263244628906, |
| "rewards/ORM": 0.6843709945678711, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.125, |
| "step": 361 |
| }, |
| { |
| "completion_length": 46.796875, |
| "epoch": 0.4008859357696567, |
| "grad_norm": 0.5340003371238708, |
| "kl": 0.03424072265625, |
| "learning_rate": 7.7375e-07, |
| "loss": -0.006476349895820022, |
| "reward": 1.877703845500946, |
| "reward_std": 0.4012419879436493, |
| "rewards/GDino": 0.7290347218513489, |
| "rewards/GIT": 0.45925381779670715, |
| "rewards/HPSv2": 0.2832450866699219, |
| "rewards/ORM": 0.4061700701713562, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.375, |
| "step": 362 |
| }, |
| { |
| "completion_length": 49.234375, |
| "epoch": 0.4019933554817276, |
| "grad_norm": 0.6717630624771118, |
| "kl": 0.0328369140625, |
| "learning_rate": 7.731249999999999e-07, |
| "loss": -0.008118146331980824, |
| "reward": 1.9765561819076538, |
| "reward_std": 0.37670013308525085, |
| "rewards/GDino": 0.6909117698669434, |
| "rewards/GIT": 0.4758438169956207, |
| "rewards/HPSv2": 0.2585926055908203, |
| "rewards/ORM": 0.5512078106403351, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -20.5625, |
| "step": 363 |
| }, |
| { |
| "completion_length": 41.453125, |
| "epoch": 0.40310077519379844, |
| "grad_norm": 0.7043358087539673, |
| "kl": 0.0404052734375, |
| "learning_rate": 7.724999999999999e-07, |
| "loss": 0.03979423549026251, |
| "reward": 2.4432953596115112, |
| "reward_std": 0.2629931718111038, |
| "rewards/GDino": 0.7979166805744171, |
| "rewards/GIT": 0.5859342813491821, |
| "rewards/HPSv2": 0.28001976013183594, |
| "rewards/ORM": 0.7794246673583984, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.5625, |
| "step": 364 |
| }, |
| { |
| "completion_length": 44.8125, |
| "epoch": 0.40420819490586934, |
| "grad_norm": 0.6534931063652039, |
| "kl": 0.055908203125, |
| "learning_rate": 7.718749999999999e-07, |
| "loss": -0.007944567129015923, |
| "reward": 1.8853873014450073, |
| "reward_std": 0.39929990470409393, |
| "rewards/GDino": 0.5773958265781403, |
| "rewards/GIT": 0.29893604665994644, |
| "rewards/HPSv2": 0.2607097625732422, |
| "rewards/ORM": 0.7483455240726471, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.25, |
| "step": 365 |
| }, |
| { |
| "completion_length": 41.546875, |
| "epoch": 0.4053156146179402, |
| "grad_norm": 0.5383189916610718, |
| "kl": 0.048095703125, |
| "learning_rate": 7.712499999999999e-07, |
| "loss": 0.013786105439066887, |
| "reward": 1.8352028131484985, |
| "reward_std": 0.4158010184764862, |
| "rewards/GDino": 0.6937500238418579, |
| "rewards/GIT": 0.19948893785476685, |
| "rewards/HPSv2": 0.26660919189453125, |
| "rewards/ORM": 0.6753546893596649, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.625, |
| "step": 366 |
| }, |
| { |
| "completion_length": 42.734375, |
| "epoch": 0.4064230343300111, |
| "grad_norm": 0.5055248141288757, |
| "kl": 0.0382080078125, |
| "learning_rate": 7.706249999999999e-07, |
| "loss": 0.01024163793772459, |
| "reward": 2.259072184562683, |
| "reward_std": 0.45819520950317383, |
| "rewards/GDino": 0.7351614236831665, |
| "rewards/GIT": 0.4725143015384674, |
| "rewards/HPSv2": 0.2722015380859375, |
| "rewards/ORM": 0.7791949510574341, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.3125, |
| "step": 367 |
| }, |
| { |
| "completion_length": 35.171875, |
| "epoch": 0.40753045404208194, |
| "grad_norm": 0.9312332272529602, |
| "kl": 0.0435791015625, |
| "learning_rate": 7.699999999999999e-07, |
| "loss": -0.02036240417510271, |
| "reward": 2.231248438358307, |
| "reward_std": 0.2120247781276703, |
| "rewards/GDino": 0.8772628009319305, |
| "rewards/GIT": 0.5012119859457016, |
| "rewards/HPSv2": 0.2806358337402344, |
| "rewards/ORM": 0.5721378028392792, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.3125, |
| "step": 368 |
| }, |
| { |
| "completion_length": 33.625, |
| "epoch": 0.40863787375415284, |
| "grad_norm": 2.0985968112945557, |
| "kl": 0.0401611328125, |
| "learning_rate": 7.69375e-07, |
| "loss": 0.0018158061429858208, |
| "reward": 2.397477388381958, |
| "reward_std": 0.40679168701171875, |
| "rewards/GDino": 0.8972452282905579, |
| "rewards/GIT": 0.629754438996315, |
| "rewards/HPSv2": 0.27672767639160156, |
| "rewards/ORM": 0.59375, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.5, |
| "step": 369 |
| }, |
| { |
| "completion_length": 44.203125, |
| "epoch": 0.4097452934662237, |
| "grad_norm": 0.6264777779579163, |
| "kl": 0.02813720703125, |
| "learning_rate": 7.6875e-07, |
| "loss": 0.022082864306867123, |
| "reward": 2.269244074821472, |
| "reward_std": 0.44469980895519257, |
| "rewards/GDino": 0.7298951148986816, |
| "rewards/GIT": 0.5341291725635529, |
| "rewards/HPSv2": 0.2665119171142578, |
| "rewards/ORM": 0.7387078404426575, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.5, |
| "step": 370 |
| }, |
| { |
| "completion_length": 38.28125, |
| "epoch": 0.4108527131782946, |
| "grad_norm": 0.6259306073188782, |
| "kl": 0.01837158203125, |
| "learning_rate": 7.68125e-07, |
| "loss": -0.0007002539932727814, |
| "reward": 2.4523496627807617, |
| "reward_std": 0.2647937461733818, |
| "rewards/GDino": 0.9088541865348816, |
| "rewards/GIT": 0.7134121358394623, |
| "rewards/HPSv2": 0.2628498077392578, |
| "rewards/ORM": 0.5672334432601929, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.8125, |
| "step": 371 |
| }, |
| { |
| "completion_length": 45.765625, |
| "epoch": 0.4119601328903654, |
| "grad_norm": 0.6793559789657593, |
| "kl": 0.0372314453125, |
| "learning_rate": 7.675e-07, |
| "loss": 0.033544646576046944, |
| "reward": 2.0182183980941772, |
| "reward_std": 0.32042600214481354, |
| "rewards/GDino": 0.6397264897823334, |
| "rewards/GIT": 0.44353990256786346, |
| "rewards/HPSv2": 0.2689342498779297, |
| "rewards/ORM": 0.6660176515579224, |
| "self_certainty_semantic": -24.875, |
| "self_certainty_token": -21.1875, |
| "step": 372 |
| }, |
| { |
| "completion_length": 42.484375, |
| "epoch": 0.4130675526024363, |
| "grad_norm": 0.5748891234397888, |
| "kl": 0.041259765625, |
| "learning_rate": 7.66875e-07, |
| "loss": -0.012200751341879368, |
| "reward": 1.907440185546875, |
| "reward_std": 0.5503653585910797, |
| "rewards/GDino": 0.7475000023841858, |
| "rewards/GIT": 0.27429039031267166, |
| "rewards/HPSv2": 0.27011680603027344, |
| "rewards/ORM": 0.6155330836772919, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.5, |
| "step": 373 |
| }, |
| { |
| "completion_length": 45.109375, |
| "epoch": 0.4141749723145072, |
| "grad_norm": 0.6625725626945496, |
| "kl": 0.03057861328125, |
| "learning_rate": 7.6625e-07, |
| "loss": -0.005671075778082013, |
| "reward": 2.1532318592071533, |
| "reward_std": 0.3619081676006317, |
| "rewards/GDino": 0.8202139139175415, |
| "rewards/GIT": 0.5140966400504112, |
| "rewards/HPSv2": 0.2699394226074219, |
| "rewards/ORM": 0.5489819049835205, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.6875, |
| "step": 374 |
| }, |
| { |
| "completion_length": 33.640625, |
| "epoch": 0.4152823920265781, |
| "grad_norm": 0.5296221375465393, |
| "kl": 0.0487060546875, |
| "learning_rate": 7.65625e-07, |
| "loss": -0.0007148468866944313, |
| "reward": 2.4465746879577637, |
| "reward_std": 0.23157501220703125, |
| "rewards/GDino": 0.8728718757629395, |
| "rewards/GIT": 0.6998476386070251, |
| "rewards/HPSv2": 0.2731456756591797, |
| "rewards/ORM": 0.6007093787193298, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.5, |
| "step": 375 |
| }, |
| { |
| "completion_length": 34.859375, |
| "epoch": 0.416389811738649, |
| "grad_norm": 0.5901670455932617, |
| "kl": 0.04443359375, |
| "learning_rate": 7.65e-07, |
| "loss": 0.0027307383716106415, |
| "reward": 2.1622787714004517, |
| "reward_std": 0.42810335755348206, |
| "rewards/GDino": 0.7800533771514893, |
| "rewards/GIT": 0.408367857336998, |
| "rewards/HPSv2": 0.25921058654785156, |
| "rewards/ORM": 0.7146469056606293, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.25, |
| "step": 376 |
| }, |
| { |
| "completion_length": 46.25, |
| "epoch": 0.4174972314507198, |
| "grad_norm": 0.6768283843994141, |
| "kl": 0.03204345703125, |
| "learning_rate": 7.64375e-07, |
| "loss": 0.021322970744222403, |
| "reward": 1.8171244859695435, |
| "reward_std": 0.18127407226711512, |
| "rewards/GDino": 0.6758842766284943, |
| "rewards/GIT": 0.3382781371474266, |
| "rewards/HPSv2": 0.2610912322998047, |
| "rewards/ORM": 0.5418707728385925, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.3125, |
| "step": 377 |
| }, |
| { |
| "completion_length": 54.15625, |
| "epoch": 0.4186046511627907, |
| "grad_norm": 0.5024600625038147, |
| "kl": 0.034912109375, |
| "learning_rate": 7.6375e-07, |
| "loss": 0.0020371037535369396, |
| "reward": 2.338524103164673, |
| "reward_std": 0.43607713282108307, |
| "rewards/GDino": 0.7437655329704285, |
| "rewards/GIT": 0.506206676363945, |
| "rewards/HPSv2": 0.27953147888183594, |
| "rewards/ORM": 0.8090203106403351, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -21.1875, |
| "step": 378 |
| }, |
| { |
| "completion_length": 39.265625, |
| "epoch": 0.41971207087486156, |
| "grad_norm": 1.9171770811080933, |
| "kl": 0.03411865234375, |
| "learning_rate": 7.63125e-07, |
| "loss": 0.002377159893512726, |
| "reward": 2.2468537092208862, |
| "reward_std": 0.20854417979717255, |
| "rewards/GDino": 0.7836942076683044, |
| "rewards/GIT": 0.5418351590633392, |
| "rewards/HPSv2": 0.260040283203125, |
| "rewards/ORM": 0.6612841784954071, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.3125, |
| "step": 379 |
| }, |
| { |
| "completion_length": 39.484375, |
| "epoch": 0.42081949058693247, |
| "grad_norm": 1.0356440544128418, |
| "kl": 0.03045654296875, |
| "learning_rate": 7.624999999999999e-07, |
| "loss": -0.014718352816998959, |
| "reward": 1.871756136417389, |
| "reward_std": 0.3412375822663307, |
| "rewards/GDino": 0.7783447802066803, |
| "rewards/GIT": 0.5043638348579407, |
| "rewards/HPSv2": 0.27927589416503906, |
| "rewards/ORM": 0.30977167189121246, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.6875, |
| "step": 380 |
| }, |
| { |
| "completion_length": 40.96875, |
| "epoch": 0.4219269102990033, |
| "grad_norm": 0.632671594619751, |
| "kl": 0.048095703125, |
| "learning_rate": 7.618749999999999e-07, |
| "loss": -0.01816024724394083, |
| "reward": 2.3965006470680237, |
| "reward_std": 0.34211961925029755, |
| "rewards/GDino": 0.8125, |
| "rewards/GIT": 0.6881425678730011, |
| "rewards/HPSv2": 0.2642993927001953, |
| "rewards/ORM": 0.631558746099472, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.0, |
| "step": 381 |
| }, |
| { |
| "completion_length": 42.890625, |
| "epoch": 0.4230343300110742, |
| "grad_norm": 0.5367975831031799, |
| "kl": 0.03173828125, |
| "learning_rate": 7.612499999999999e-07, |
| "loss": 0.005420586094260216, |
| "reward": 2.1896389722824097, |
| "reward_std": 0.3109910786151886, |
| "rewards/GDino": 0.7874999940395355, |
| "rewards/GIT": 0.49838483333587646, |
| "rewards/HPSv2": 0.27602577209472656, |
| "rewards/ORM": 0.6277283728122711, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -20.75, |
| "step": 382 |
| }, |
| { |
| "completion_length": 42.171875, |
| "epoch": 0.42414174972314506, |
| "grad_norm": 0.3789452910423279, |
| "kl": 0.03533935546875, |
| "learning_rate": 7.606249999999999e-07, |
| "loss": 0.014071432873606682, |
| "reward": 1.9684094190597534, |
| "reward_std": 0.5238681137561798, |
| "rewards/GDino": 0.7198299765586853, |
| "rewards/GIT": 0.5013139694929123, |
| "rewards/HPSv2": 0.2457408905029297, |
| "rewards/ORM": 0.5015245378017426, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.25, |
| "step": 383 |
| }, |
| { |
| "completion_length": 40.90625, |
| "epoch": 0.42524916943521596, |
| "grad_norm": 2.8351011276245117, |
| "kl": 0.0416259765625, |
| "learning_rate": 7.599999999999999e-07, |
| "loss": -0.021148506551980972, |
| "reward": 2.102011501789093, |
| "reward_std": 0.2582581639289856, |
| "rewards/GDino": 0.7812051475048065, |
| "rewards/GIT": 0.5529660955071449, |
| "rewards/HPSv2": 0.27136993408203125, |
| "rewards/ORM": 0.4964703768491745, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.25, |
| "step": 384 |
| }, |
| { |
| "completion_length": 51.046875, |
| "epoch": 0.4263565891472868, |
| "grad_norm": 0.6955005526542664, |
| "kl": 0.0413818359375, |
| "learning_rate": 7.59375e-07, |
| "loss": 0.0069068125449121, |
| "reward": 2.098905622959137, |
| "reward_std": 0.2885952666401863, |
| "rewards/GDino": 0.7392346262931824, |
| "rewards/GIT": 0.32319432497024536, |
| "rewards/HPSv2": 0.26218605041503906, |
| "rewards/ORM": 0.7742906510829926, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -20.875, |
| "step": 385 |
| }, |
| { |
| "completion_length": 29.90625, |
| "epoch": 0.4274640088593577, |
| "grad_norm": 0.8517731428146362, |
| "kl": 0.0516357421875, |
| "learning_rate": 7.5875e-07, |
| "loss": 0.006788903381675482, |
| "reward": 2.4118454456329346, |
| "reward_std": 0.3538888171315193, |
| "rewards/GDino": 0.785937488079071, |
| "rewards/GIT": 0.5628680139780045, |
| "rewards/HPSv2": 0.28178977966308594, |
| "rewards/ORM": 0.78125, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.4375, |
| "step": 386 |
| }, |
| { |
| "completion_length": 36.765625, |
| "epoch": 0.42857142857142855, |
| "grad_norm": 0.6707489490509033, |
| "kl": 0.0411376953125, |
| "learning_rate": 7.58125e-07, |
| "loss": 0.014098147861659527, |
| "reward": 2.149623155593872, |
| "reward_std": 0.42522846162319183, |
| "rewards/GDino": 0.8425147533416748, |
| "rewards/GIT": 0.49068495631217957, |
| "rewards/HPSv2": 0.26511573791503906, |
| "rewards/ORM": 0.5513076782226562, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.5625, |
| "step": 387 |
| }, |
| { |
| "completion_length": 37.625, |
| "epoch": 0.42967884828349945, |
| "grad_norm": 0.4950082004070282, |
| "kl": 0.0244140625, |
| "learning_rate": 7.575e-07, |
| "loss": -0.012308157398365438, |
| "reward": 2.5664467811584473, |
| "reward_std": 0.319002628326416, |
| "rewards/GDino": 0.8304032683372498, |
| "rewards/GIT": 0.6985468864440918, |
| "rewards/HPSv2": 0.2795543670654297, |
| "rewards/ORM": 0.7579423785209656, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.125, |
| "step": 388 |
| }, |
| { |
| "completion_length": 41.625, |
| "epoch": 0.43078626799557035, |
| "grad_norm": 0.7242435216903687, |
| "kl": 0.0421142578125, |
| "learning_rate": 7.56875e-07, |
| "loss": -0.0010866057127714157, |
| "reward": 2.184369921684265, |
| "reward_std": 0.47019390761852264, |
| "rewards/GDino": 0.7906249761581421, |
| "rewards/GIT": 0.46294473111629486, |
| "rewards/HPSv2": 0.2710704803466797, |
| "rewards/ORM": 0.6597297042608261, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.625, |
| "step": 389 |
| }, |
| { |
| "completion_length": 50.125, |
| "epoch": 0.4318936877076412, |
| "grad_norm": 1.1640534400939941, |
| "kl": 0.030517578125, |
| "learning_rate": 7.5625e-07, |
| "loss": -0.019844084046781063, |
| "reward": 2.396089255809784, |
| "reward_std": 0.24368739873170853, |
| "rewards/GDino": 0.7795874178409576, |
| "rewards/GIT": 0.6525524854660034, |
| "rewards/HPSv2": 0.26082420349121094, |
| "rewards/ORM": 0.703125, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.25, |
| "step": 390 |
| }, |
| { |
| "completion_length": 35.984375, |
| "epoch": 0.4330011074197121, |
| "grad_norm": 0.5103877782821655, |
| "kl": 0.0311279296875, |
| "learning_rate": 7.55625e-07, |
| "loss": -0.008262321585789323, |
| "reward": 2.406006336212158, |
| "reward_std": 0.23674829304218292, |
| "rewards/GDino": 0.8994598388671875, |
| "rewards/GIT": 0.6523403823375702, |
| "rewards/HPSv2": 0.2604560852050781, |
| "rewards/ORM": 0.5937500149011612, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.9375, |
| "step": 391 |
| }, |
| { |
| "completion_length": 43.921875, |
| "epoch": 0.43410852713178294, |
| "grad_norm": 0.497820645570755, |
| "kl": 0.042724609375, |
| "learning_rate": 7.55e-07, |
| "loss": -0.00259082525735721, |
| "reward": 1.8813644647598267, |
| "reward_std": 0.4379560351371765, |
| "rewards/GDino": 0.7333734929561615, |
| "rewards/GIT": 0.3741501718759537, |
| "rewards/HPSv2": 0.2706317901611328, |
| "rewards/ORM": 0.5032089352607727, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.5, |
| "step": 392 |
| }, |
| { |
| "completion_length": 30.484375, |
| "epoch": 0.43521594684385384, |
| "grad_norm": 0.6349219083786011, |
| "kl": 0.0517578125, |
| "learning_rate": 7.54375e-07, |
| "loss": -0.013653552625328302, |
| "reward": 2.2154151797294617, |
| "reward_std": 0.2131977081298828, |
| "rewards/GDino": 0.8221572637557983, |
| "rewards/GIT": 0.4915757179260254, |
| "rewards/HPSv2": 0.28302001953125, |
| "rewards/ORM": 0.6186620593070984, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.8125, |
| "step": 393 |
| }, |
| { |
| "completion_length": 39.859375, |
| "epoch": 0.4363233665559247, |
| "grad_norm": 0.569675087928772, |
| "kl": 0.03436279296875, |
| "learning_rate": 7.5375e-07, |
| "loss": -0.006760713644325733, |
| "reward": 2.4199535846710205, |
| "reward_std": 0.302301824092865, |
| "rewards/GDino": 0.9265625178813934, |
| "rewards/GIT": 0.4195697158575058, |
| "rewards/HPSv2": 0.264801025390625, |
| "rewards/ORM": 0.8090203106403351, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.4375, |
| "step": 394 |
| }, |
| { |
| "completion_length": 38.171875, |
| "epoch": 0.4374307862679956, |
| "grad_norm": 1.2075111865997314, |
| "kl": 0.0443115234375, |
| "learning_rate": 7.53125e-07, |
| "loss": 0.009720378555357456, |
| "reward": 2.379997491836548, |
| "reward_std": 0.34344032406806946, |
| "rewards/GDino": 0.8730711340904236, |
| "rewards/GIT": 0.5321757197380066, |
| "rewards/HPSv2": 0.2716255187988281, |
| "rewards/ORM": 0.7031249701976776, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.5625, |
| "step": 395 |
| }, |
| { |
| "completion_length": 38.6875, |
| "epoch": 0.43853820598006643, |
| "grad_norm": 342.09454345703125, |
| "kl": 5.0242919921875, |
| "learning_rate": 7.524999999999999e-07, |
| "loss": 0.027052484452724457, |
| "reward": 2.2779553532600403, |
| "reward_std": 0.4027100205421448, |
| "rewards/GDino": 0.8293097615242004, |
| "rewards/GIT": 0.4535738229751587, |
| "rewards/HPSv2": 0.2744464874267578, |
| "rewards/ORM": 0.7206252217292786, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.25, |
| "step": 396 |
| }, |
| { |
| "completion_length": 38.515625, |
| "epoch": 0.43964562569213733, |
| "grad_norm": 0.5501837730407715, |
| "kl": 0.03302001953125, |
| "learning_rate": 7.518749999999999e-07, |
| "loss": -0.02949504740536213, |
| "reward": 2.547436833381653, |
| "reward_std": 0.37221017479896545, |
| "rewards/GDino": 0.928906261920929, |
| "rewards/GIT": 0.6719013452529907, |
| "rewards/HPSv2": 0.2500629425048828, |
| "rewards/ORM": 0.696566253900528, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.625, |
| "step": 397 |
| }, |
| { |
| "completion_length": 36.8125, |
| "epoch": 0.4407530454042082, |
| "grad_norm": 1.14122474193573, |
| "kl": 0.0433349609375, |
| "learning_rate": 7.512499999999999e-07, |
| "loss": 0.0011832071468234062, |
| "reward": 2.6030901670455933, |
| "reward_std": 0.367518350481987, |
| "rewards/GDino": 0.8026041388511658, |
| "rewards/GIT": 0.6320919096469879, |
| "rewards/HPSv2": 0.2777690887451172, |
| "rewards/ORM": 0.890625, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.1875, |
| "step": 398 |
| }, |
| { |
| "completion_length": 46.546875, |
| "epoch": 0.4418604651162791, |
| "grad_norm": 0.5975333452224731, |
| "kl": 0.04150390625, |
| "learning_rate": 7.506249999999999e-07, |
| "loss": 0.016775197349488735, |
| "reward": 1.8540863394737244, |
| "reward_std": 0.29581695050001144, |
| "rewards/GDino": 0.6284334063529968, |
| "rewards/GIT": 0.23308932036161423, |
| "rewards/HPSv2": 0.25658416748046875, |
| "rewards/ORM": 0.7359794676303864, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.5625, |
| "step": 399 |
| }, |
| { |
| "completion_length": 36.34375, |
| "epoch": 0.4429678848283499, |
| "grad_norm": 1.3476744890213013, |
| "kl": 0.0401611328125, |
| "learning_rate": 7.5e-07, |
| "loss": -0.013706179801374674, |
| "reward": 2.096378445625305, |
| "reward_std": 0.362843781709671, |
| "rewards/GDino": 0.7787023782730103, |
| "rewards/GIT": 0.4101857915520668, |
| "rewards/HPSv2": 0.27315330505371094, |
| "rewards/ORM": 0.6343370079994202, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.0625, |
| "step": 400 |
| }, |
| { |
| "completion_length": 34.140625, |
| "epoch": 0.4440753045404208, |
| "grad_norm": 0.7070680260658264, |
| "kl": 0.03662109375, |
| "learning_rate": 7.49375e-07, |
| "loss": -0.017339009791612625, |
| "reward": 2.837607264518738, |
| "reward_std": 0.2649471387267113, |
| "rewards/GDino": 0.971875011920929, |
| "rewards/GIT": 0.7277626097202301, |
| "rewards/HPSv2": 0.2670707702636719, |
| "rewards/ORM": 0.8708988130092621, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.4375, |
| "step": 401 |
| }, |
| { |
| "completion_length": 36.9375, |
| "epoch": 0.44518272425249167, |
| "grad_norm": 0.466122567653656, |
| "kl": 0.038818359375, |
| "learning_rate": 7.4875e-07, |
| "loss": -0.00474184425547719, |
| "reward": 2.21277391910553, |
| "reward_std": 0.2819427028298378, |
| "rewards/GDino": 0.7564797103404999, |
| "rewards/GIT": 0.4613865911960602, |
| "rewards/HPSv2": 0.2831687927246094, |
| "rewards/ORM": 0.7117387652397156, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.4375, |
| "step": 402 |
| }, |
| { |
| "completion_length": 41.96875, |
| "epoch": 0.44629014396456257, |
| "grad_norm": 0.6223385334014893, |
| "kl": 0.041015625, |
| "learning_rate": 7.48125e-07, |
| "loss": 0.006069277413189411, |
| "reward": 1.8274257183074951, |
| "reward_std": 0.277290478348732, |
| "rewards/GDino": 0.5828124582767487, |
| "rewards/GIT": 0.23541270196437836, |
| "rewards/HPSv2": 0.2838916778564453, |
| "rewards/ORM": 0.725308746099472, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.75, |
| "step": 403 |
| }, |
| { |
| "completion_length": 41.828125, |
| "epoch": 0.44739756367663347, |
| "grad_norm": 0.8799866437911987, |
| "kl": 0.045654296875, |
| "learning_rate": 7.475e-07, |
| "loss": 0.0017385446117259562, |
| "reward": 1.9244996309280396, |
| "reward_std": 0.3593953996896744, |
| "rewards/GDino": 0.6904172003269196, |
| "rewards/GIT": 0.3588860034942627, |
| "rewards/HPSv2": 0.2580089569091797, |
| "rewards/ORM": 0.6171875, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0, |
| "step": 404 |
| }, |
| { |
| "completion_length": 49.75, |
| "epoch": 0.4485049833887043, |
| "grad_norm": 1.2903233766555786, |
| "kl": 0.0443115234375, |
| "learning_rate": 7.468749999999999e-07, |
| "loss": 0.004371013026684523, |
| "reward": 2.7583900690078735, |
| "reward_std": 0.2406376302242279, |
| "rewards/GDino": 0.8804058730602264, |
| "rewards/GIT": 0.6657536029815674, |
| "rewards/HPSv2": 0.2673187255859375, |
| "rewards/ORM": 0.9449118673801422, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.4375, |
| "step": 405 |
| }, |
| { |
| "completion_length": 43.109375, |
| "epoch": 0.4496124031007752, |
| "grad_norm": 1.4691380262374878, |
| "kl": 0.046630859375, |
| "learning_rate": 7.4625e-07, |
| "loss": 0.016827881336212158, |
| "reward": 1.6681452989578247, |
| "reward_std": 0.37892957031726837, |
| "rewards/GDino": 0.6188920736312866, |
| "rewards/GIT": 0.03339381515979767, |
| "rewards/HPSv2": 0.282958984375, |
| "rewards/ORM": 0.7329003810882568, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.375, |
| "step": 406 |
| }, |
| { |
| "completion_length": 38.0, |
| "epoch": 0.45071982281284606, |
| "grad_norm": 0.7037851810455322, |
| "kl": 0.0367431640625, |
| "learning_rate": 7.45625e-07, |
| "loss": -0.017478429013863206, |
| "reward": 1.9953652024269104, |
| "reward_std": 0.364028662443161, |
| "rewards/GDino": 0.7403169274330139, |
| "rewards/GIT": 0.406859390437603, |
| "rewards/HPSv2": 0.2685394287109375, |
| "rewards/ORM": 0.5796495079994202, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -22.25, |
| "step": 407 |
| }, |
| { |
| "completion_length": 44.0625, |
| "epoch": 0.45182724252491696, |
| "grad_norm": 0.5131052136421204, |
| "kl": 0.0306396484375, |
| "learning_rate": 7.45e-07, |
| "loss": 0.005040338612161577, |
| "reward": 2.041200637817383, |
| "reward_std": 0.3807089328765869, |
| "rewards/GDino": 0.7476771473884583, |
| "rewards/GIT": 0.4045151174068451, |
| "rewards/HPSv2": 0.25945472717285156, |
| "rewards/ORM": 0.6295536458492279, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.125, |
| "step": 408 |
| }, |
| { |
| "completion_length": 38.34375, |
| "epoch": 0.4529346622369878, |
| "grad_norm": 0.636464536190033, |
| "kl": 0.0491943359375, |
| "learning_rate": 7.44375e-07, |
| "loss": -0.03190365340560675, |
| "reward": 2.7317603826522827, |
| "reward_std": 0.4104766994714737, |
| "rewards/GDino": 0.9453125, |
| "rewards/GIT": 0.7198237776756287, |
| "rewards/HPSv2": 0.27568626403808594, |
| "rewards/ORM": 0.7909377217292786, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -20.9375, |
| "step": 409 |
| }, |
| { |
| "completion_length": 33.546875, |
| "epoch": 0.4540420819490587, |
| "grad_norm": 0.8647944927215576, |
| "kl": 0.042724609375, |
| "learning_rate": 7.4375e-07, |
| "loss": -0.003960296045988798, |
| "reward": 2.0273808240890503, |
| "reward_std": 0.3602750897407532, |
| "rewards/GDino": 0.8088895082473755, |
| "rewards/GIT": 0.4431849420070648, |
| "rewards/HPSv2": 0.2676239013671875, |
| "rewards/ORM": 0.5076826512813568, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.375, |
| "step": 410 |
| }, |
| { |
| "completion_length": 42.140625, |
| "epoch": 0.45514950166112955, |
| "grad_norm": 0.4983558654785156, |
| "kl": 0.0391845703125, |
| "learning_rate": 7.43125e-07, |
| "loss": -0.019663702347315848, |
| "reward": 2.568248152732849, |
| "reward_std": 0.22275355458259583, |
| "rewards/GDino": 0.9209606051445007, |
| "rewards/GIT": 0.6949225664138794, |
| "rewards/HPSv2": 0.2756767272949219, |
| "rewards/ORM": 0.6766883730888367, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.8125, |
| "step": 411 |
| }, |
| { |
| "completion_length": 36.234375, |
| "epoch": 0.45625692137320045, |
| "grad_norm": 2.719592332839966, |
| "kl": 0.0440673828125, |
| "learning_rate": 7.425e-07, |
| "loss": -0.012157580815255642, |
| "reward": 2.434681475162506, |
| "reward_std": 0.13174384832382202, |
| "rewards/GDino": 0.8247999548912048, |
| "rewards/GIT": 0.46177390962839127, |
| "rewards/HPSv2": 0.27869415283203125, |
| "rewards/ORM": 0.869413435459137, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.0, |
| "step": 412 |
| }, |
| { |
| "completion_length": 36.046875, |
| "epoch": 0.4573643410852713, |
| "grad_norm": 0.5472164750099182, |
| "kl": 0.0443115234375, |
| "learning_rate": 7.418749999999999e-07, |
| "loss": 0.001380805391818285, |
| "reward": 1.6478554010391235, |
| "reward_std": 0.39775100350379944, |
| "rewards/GDino": 0.739335834980011, |
| "rewards/GIT": 0.12186351418495178, |
| "rewards/HPSv2": 0.271331787109375, |
| "rewards/ORM": 0.5153242349624634, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.0, |
| "step": 413 |
| }, |
| { |
| "completion_length": 42.03125, |
| "epoch": 0.4584717607973422, |
| "grad_norm": 0.9818385243415833, |
| "kl": 0.045654296875, |
| "learning_rate": 7.412499999999999e-07, |
| "loss": 0.00032033328898251057, |
| "reward": 2.244621515274048, |
| "reward_std": 0.3542056977748871, |
| "rewards/GDino": 0.7868304252624512, |
| "rewards/GIT": 0.3334681913256645, |
| "rewards/HPSv2": 0.28057289123535156, |
| "rewards/ORM": 0.84375, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.75, |
| "step": 414 |
| }, |
| { |
| "completion_length": 44.421875, |
| "epoch": 0.45957918050941304, |
| "grad_norm": 1.3210840225219727, |
| "kl": 0.043212890625, |
| "learning_rate": 7.406249999999999e-07, |
| "loss": 0.01814082730561495, |
| "reward": 2.018928349018097, |
| "reward_std": 0.5640377998352051, |
| "rewards/GDino": 0.7119726538658142, |
| "rewards/GIT": 0.3103228211402893, |
| "rewards/HPSv2": 0.26573753356933594, |
| "rewards/ORM": 0.7308953106403351, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.1875, |
| "step": 415 |
| }, |
| { |
| "completion_length": 37.140625, |
| "epoch": 0.46068660022148394, |
| "grad_norm": 0.6978505253791809, |
| "kl": 0.040283203125, |
| "learning_rate": 7.4e-07, |
| "loss": 0.02266329526901245, |
| "reward": 1.9420585632324219, |
| "reward_std": 0.22528188675642014, |
| "rewards/GDino": 0.6741819083690643, |
| "rewards/GIT": 0.326813742518425, |
| "rewards/HPSv2": 0.25824642181396484, |
| "rewards/ORM": 0.6828164756298065, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.375, |
| "step": 416 |
| }, |
| { |
| "completion_length": 47.140625, |
| "epoch": 0.46179401993355484, |
| "grad_norm": 0.6955863833427429, |
| "kl": 0.02899169921875, |
| "learning_rate": 7.39375e-07, |
| "loss": -0.022133303806185722, |
| "reward": 2.2546093463897705, |
| "reward_std": 0.37237268686294556, |
| "rewards/GDino": 0.8068858683109283, |
| "rewards/GIT": 0.51471146941185, |
| "rewards/HPSv2": 0.25958251953125, |
| "rewards/ORM": 0.6734296083450317, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.4375, |
| "step": 417 |
| }, |
| { |
| "completion_length": 36.25, |
| "epoch": 0.4629014396456257, |
| "grad_norm": 0.6875625252723694, |
| "kl": 0.0550537109375, |
| "learning_rate": 7.3875e-07, |
| "loss": -0.02608813438564539, |
| "reward": 1.9164315462112427, |
| "reward_std": 0.4128989577293396, |
| "rewards/GDino": 0.737500011920929, |
| "rewards/GIT": 0.48389841616153717, |
| "rewards/HPSv2": 0.25507545471191406, |
| "rewards/ORM": 0.4399575889110565, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.8125, |
| "step": 418 |
| }, |
| { |
| "completion_length": 42.359375, |
| "epoch": 0.4640088593576966, |
| "grad_norm": 0.5099749565124512, |
| "kl": 0.035888671875, |
| "learning_rate": 7.38125e-07, |
| "loss": 0.012332713231444359, |
| "reward": 2.1047881841659546, |
| "reward_std": 0.3690176010131836, |
| "rewards/GDino": 0.6977547407150269, |
| "rewards/GIT": 0.5606772899627686, |
| "rewards/HPSv2": 0.24872398376464844, |
| "rewards/ORM": 0.5976322144269943, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.875, |
| "step": 419 |
| }, |
| { |
| "completion_length": 47.796875, |
| "epoch": 0.46511627906976744, |
| "grad_norm": 0.7968041896820068, |
| "kl": 0.02825927734375, |
| "learning_rate": 7.375e-07, |
| "loss": -0.007890310138463974, |
| "reward": 2.5399848222732544, |
| "reward_std": 0.2056763395667076, |
| "rewards/GDino": 0.9003763794898987, |
| "rewards/GIT": 0.6921442151069641, |
| "rewards/HPSv2": 0.28192710876464844, |
| "rewards/ORM": 0.6655370742082596, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -22.1875, |
| "step": 420 |
| }, |
| { |
| "completion_length": 40.078125, |
| "epoch": 0.46622369878183834, |
| "grad_norm": 0.5168143510818481, |
| "kl": 0.039794921875, |
| "learning_rate": 7.368749999999999e-07, |
| "loss": 7.763970643281937e-06, |
| "reward": 2.2306275367736816, |
| "reward_std": 0.19530287384986877, |
| "rewards/GDino": 0.7578124701976776, |
| "rewards/GIT": 0.5601051151752472, |
| "rewards/HPSv2": 0.2732887268066406, |
| "rewards/ORM": 0.6394211947917938, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.9375, |
| "step": 421 |
| }, |
| { |
| "completion_length": 47.09375, |
| "epoch": 0.4673311184939092, |
| "grad_norm": 0.6787537932395935, |
| "kl": 0.0335693359375, |
| "learning_rate": 7.362499999999999e-07, |
| "loss": 0.023679735139012337, |
| "reward": 2.0749627351760864, |
| "reward_std": 0.2599632367491722, |
| "rewards/GDino": 0.7587704658508301, |
| "rewards/GIT": 0.5343009531497955, |
| "rewards/HPSv2": 0.2624359130859375, |
| "rewards/ORM": 0.5194553732872009, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -22.1875, |
| "step": 422 |
| }, |
| { |
| "completion_length": 45.15625, |
| "epoch": 0.4684385382059801, |
| "grad_norm": 0.5317529439926147, |
| "kl": 0.03717041015625, |
| "learning_rate": 7.356249999999999e-07, |
| "loss": 0.0031719468533992767, |
| "reward": 2.253452777862549, |
| "reward_std": 0.2863696217536926, |
| "rewards/GDino": 0.8162786066532135, |
| "rewards/GIT": 0.5946633368730545, |
| "rewards/HPSv2": 0.2769317626953125, |
| "rewards/ORM": 0.5655790567398071, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.375, |
| "step": 423 |
| }, |
| { |
| "completion_length": 43.578125, |
| "epoch": 0.4695459579180509, |
| "grad_norm": 1.1308730840682983, |
| "kl": 0.033447265625, |
| "learning_rate": 7.35e-07, |
| "loss": -0.0021982135949656367, |
| "reward": 1.969843864440918, |
| "reward_std": 0.39374539256095886, |
| "rewards/GDino": 0.7518229186534882, |
| "rewards/GIT": 0.3873000741004944, |
| "rewards/HPSv2": 0.2738075256347656, |
| "rewards/ORM": 0.556913435459137, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.75, |
| "step": 424 |
| }, |
| { |
| "completion_length": 44.140625, |
| "epoch": 0.4706533776301218, |
| "grad_norm": 0.9367016553878784, |
| "kl": 0.0478515625, |
| "learning_rate": 7.34375e-07, |
| "loss": 0.0006617754697799683, |
| "reward": 1.942332684993744, |
| "reward_std": 0.3015865758061409, |
| "rewards/GDino": 0.7671928405761719, |
| "rewards/GIT": 0.37167659401893616, |
| "rewards/HPSv2": 0.27842140197753906, |
| "rewards/ORM": 0.5250419676303864, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.375, |
| "step": 425 |
| }, |
| { |
| "completion_length": 39.921875, |
| "epoch": 0.4717607973421927, |
| "grad_norm": 0.5315147638320923, |
| "kl": 0.03662109375, |
| "learning_rate": 7.3375e-07, |
| "loss": 0.010861005634069443, |
| "reward": 1.9369670152664185, |
| "reward_std": 0.2872409373521805, |
| "rewards/GDino": 0.7253109216690063, |
| "rewards/GIT": 0.3525398448109627, |
| "rewards/HPSv2": 0.2622871398925781, |
| "rewards/ORM": 0.5968290567398071, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.9375, |
| "step": 426 |
| }, |
| { |
| "completion_length": 49.578125, |
| "epoch": 0.4728682170542636, |
| "grad_norm": 1.4368230104446411, |
| "kl": 0.02911376953125, |
| "learning_rate": 7.33125e-07, |
| "loss": 0.030126910656690598, |
| "reward": 2.7364684343338013, |
| "reward_std": 0.26522858440876007, |
| "rewards/GDino": 0.953125, |
| "rewards/GIT": 0.9096876680850983, |
| "rewards/HPSv2": 0.2548637390136719, |
| "rewards/ORM": 0.618791937828064, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.125, |
| "step": 427 |
| }, |
| { |
| "completion_length": 31.109375, |
| "epoch": 0.4739756367663344, |
| "grad_norm": 0.8413992524147034, |
| "kl": 0.036865234375, |
| "learning_rate": 7.325e-07, |
| "loss": 0.0009733177721500397, |
| "reward": 2.0109975337982178, |
| "reward_std": 0.3149056136608124, |
| "rewards/GDino": 0.8508215546607971, |
| "rewards/GIT": 0.31817829608917236, |
| "rewards/HPSv2": 0.2774925231933594, |
| "rewards/ORM": 0.5645051002502441, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.5625, |
| "step": 428 |
| }, |
| { |
| "completion_length": 38.0625, |
| "epoch": 0.4750830564784053, |
| "grad_norm": 0.5556603074073792, |
| "kl": 0.055908203125, |
| "learning_rate": 7.31875e-07, |
| "loss": -0.01675282884389162, |
| "reward": 1.731520414352417, |
| "reward_std": 0.5266979932785034, |
| "rewards/GDino": 0.6363127827644348, |
| "rewards/GIT": 0.21719177067279816, |
| "rewards/HPSv2": 0.26899147033691406, |
| "rewards/ORM": 0.6090243458747864, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.25, |
| "step": 429 |
| }, |
| { |
| "completion_length": 48.203125, |
| "epoch": 0.47619047619047616, |
| "grad_norm": 0.5894951820373535, |
| "kl": 0.052734375, |
| "learning_rate": 7.312499999999999e-07, |
| "loss": 0.027036309242248535, |
| "reward": 1.872725009918213, |
| "reward_std": 0.39114080369472504, |
| "rewards/GDino": 0.6521097123622894, |
| "rewards/GIT": 0.2845214605331421, |
| "rewards/HPSv2": 0.2642688751220703, |
| "rewards/ORM": 0.6718250513076782, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.875, |
| "step": 430 |
| }, |
| { |
| "completion_length": 55.046875, |
| "epoch": 0.47729789590254706, |
| "grad_norm": 0.5126261711120605, |
| "kl": 0.04833984375, |
| "learning_rate": 7.306249999999999e-07, |
| "loss": -0.004950821399688721, |
| "reward": 2.2202555537223816, |
| "reward_std": 0.2636689841747284, |
| "rewards/GDino": 0.7635050415992737, |
| "rewards/GIT": 0.5405832678079605, |
| "rewards/HPSv2": 0.27193260192871094, |
| "rewards/ORM": 0.6442345380783081, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.0625, |
| "step": 431 |
| }, |
| { |
| "completion_length": 42.40625, |
| "epoch": 0.47840531561461797, |
| "grad_norm": 0.8860141038894653, |
| "kl": 0.039794921875, |
| "learning_rate": 7.3e-07, |
| "loss": 0.005071769934147596, |
| "reward": 1.8272140622138977, |
| "reward_std": 0.2828010469675064, |
| "rewards/GDino": 0.6431768834590912, |
| "rewards/GIT": 0.29858846962451935, |
| "rewards/HPSv2": 0.252716064453125, |
| "rewards/ORM": 0.632732629776001, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.875, |
| "step": 432 |
| }, |
| { |
| "completion_length": 38.375, |
| "epoch": 0.4795127353266888, |
| "grad_norm": 1.3662358522415161, |
| "kl": 0.0472412109375, |
| "learning_rate": 7.29375e-07, |
| "loss": 0.0014977240934967995, |
| "reward": 2.062899112701416, |
| "reward_std": 0.5158149152994156, |
| "rewards/GDino": 0.7304880619049072, |
| "rewards/GIT": 0.3485347479581833, |
| "rewards/HPSv2": 0.2774925231933594, |
| "rewards/ORM": 0.7063838839530945, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0625, |
| "step": 433 |
| }, |
| { |
| "completion_length": 43.515625, |
| "epoch": 0.4806201550387597, |
| "grad_norm": 0.49138492345809937, |
| "kl": 0.0408935546875, |
| "learning_rate": 7.2875e-07, |
| "loss": -0.023660541512072086, |
| "reward": 2.264481842517853, |
| "reward_std": 0.3346661627292633, |
| "rewards/GDino": 0.8219202160835266, |
| "rewards/GIT": 0.5395843833684921, |
| "rewards/HPSv2": 0.2608776092529297, |
| "rewards/ORM": 0.6420996189117432, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.375, |
| "step": 434 |
| }, |
| { |
| "completion_length": 52.78125, |
| "epoch": 0.48172757475083056, |
| "grad_norm": 0.7937164902687073, |
| "kl": 0.0439453125, |
| "learning_rate": 7.28125e-07, |
| "loss": -0.005721581168472767, |
| "reward": 1.99281644821167, |
| "reward_std": 0.4047849476337433, |
| "rewards/GDino": 0.8185839653015137, |
| "rewards/GIT": 0.39505699276924133, |
| "rewards/HPSv2": 0.2604713439941406, |
| "rewards/ORM": 0.5187040269374847, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.6875, |
| "step": 435 |
| }, |
| { |
| "completion_length": 42.078125, |
| "epoch": 0.48283499446290146, |
| "grad_norm": 1.5256606340408325, |
| "kl": 0.0430908203125, |
| "learning_rate": 7.275e-07, |
| "loss": -0.008313903585076332, |
| "reward": 2.0975226163864136, |
| "reward_std": 0.3734524995088577, |
| "rewards/GDino": 0.7363665997982025, |
| "rewards/GIT": 0.4460279196500778, |
| "rewards/HPSv2": 0.2857952117919922, |
| "rewards/ORM": 0.6293327808380127, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.0625, |
| "step": 436 |
| }, |
| { |
| "completion_length": 50.0625, |
| "epoch": 0.4839424141749723, |
| "grad_norm": 0.49155765771865845, |
| "kl": 0.0369873046875, |
| "learning_rate": 7.26875e-07, |
| "loss": 0.021884786896407604, |
| "reward": 2.283669114112854, |
| "reward_std": 0.3140418156981468, |
| "rewards/GDino": 0.8053280115127563, |
| "rewards/GIT": 0.6532137989997864, |
| "rewards/HPSv2": 0.25914764404296875, |
| "rewards/ORM": 0.5659796893596649, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.1875, |
| "step": 437 |
| }, |
| { |
| "completion_length": 47.6875, |
| "epoch": 0.4850498338870432, |
| "grad_norm": 0.6744347810745239, |
| "kl": 0.0421142578125, |
| "learning_rate": 7.262499999999999e-07, |
| "loss": -0.01202036626636982, |
| "reward": 2.3710286617279053, |
| "reward_std": 0.2711103931069374, |
| "rewards/GDino": 0.7963541746139526, |
| "rewards/GIT": 0.6432071924209595, |
| "rewards/HPSv2": 0.25798797607421875, |
| "rewards/ORM": 0.6734794974327087, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.875, |
| "step": 438 |
| }, |
| { |
| "completion_length": 40.703125, |
| "epoch": 0.48615725359911405, |
| "grad_norm": 0.4731282591819763, |
| "kl": 0.0450439453125, |
| "learning_rate": 7.256249999999999e-07, |
| "loss": -0.0025211619213223457, |
| "reward": 2.326407551765442, |
| "reward_std": 0.3486577272415161, |
| "rewards/GDino": 0.8489097654819489, |
| "rewards/GIT": 0.5543791800737381, |
| "rewards/HPSv2": 0.2765064239501953, |
| "rewards/ORM": 0.6466122269630432, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.6875, |
| "step": 439 |
| }, |
| { |
| "completion_length": 42.25, |
| "epoch": 0.48726467331118495, |
| "grad_norm": 1.9416545629501343, |
| "kl": 0.0430908203125, |
| "learning_rate": 7.249999999999999e-07, |
| "loss": 0.026762551395222545, |
| "reward": 2.0378406643867493, |
| "reward_std": 0.3052428662776947, |
| "rewards/GDino": 0.7424247860908508, |
| "rewards/GIT": 0.3270244523882866, |
| "rewards/HPSv2": 0.28089141845703125, |
| "rewards/ORM": 0.6875, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.3125, |
| "step": 440 |
| }, |
| { |
| "completion_length": 38.78125, |
| "epoch": 0.4883720930232558, |
| "grad_norm": 0.5084402561187744, |
| "kl": 0.046875, |
| "learning_rate": 7.243749999999999e-07, |
| "loss": -0.001582828350365162, |
| "reward": 2.3616641759872437, |
| "reward_std": 0.35844916850328445, |
| "rewards/GDino": 0.8143379390239716, |
| "rewards/GIT": 0.5581052601337433, |
| "rewards/HPSv2": 0.27395057678222656, |
| "rewards/ORM": 0.7152703106403351, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.25, |
| "step": 441 |
| }, |
| { |
| "completion_length": 44.125, |
| "epoch": 0.4894795127353267, |
| "grad_norm": 0.8177931904792786, |
| "kl": 0.0474853515625, |
| "learning_rate": 7.2375e-07, |
| "loss": 0.005483056418597698, |
| "reward": 1.7684773802757263, |
| "reward_std": 0.5146408230066299, |
| "rewards/GDino": 0.664166659116745, |
| "rewards/GIT": 0.25258100032806396, |
| "rewards/HPSv2": 0.2692718505859375, |
| "rewards/ORM": 0.5824578404426575, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.75, |
| "step": 442 |
| }, |
| { |
| "completion_length": 35.5, |
| "epoch": 0.49058693244739754, |
| "grad_norm": 0.6697931885719299, |
| "kl": 0.0384521484375, |
| "learning_rate": 7.23125e-07, |
| "loss": -0.001371758058667183, |
| "reward": 1.694158136844635, |
| "reward_std": 0.26968376338481903, |
| "rewards/GDino": 0.6942708194255829, |
| "rewards/GIT": 0.43527622520923615, |
| "rewards/HPSv2": 0.24617385864257812, |
| "rewards/ORM": 0.31843727827072144, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.875, |
| "step": 443 |
| }, |
| { |
| "completion_length": 43.078125, |
| "epoch": 0.49169435215946844, |
| "grad_norm": 0.6402589082717896, |
| "kl": 0.0433349609375, |
| "learning_rate": 7.225e-07, |
| "loss": 0.03987434785813093, |
| "reward": 1.8952922821044922, |
| "reward_std": 0.36800456047058105, |
| "rewards/GDino": 0.7403267621994019, |
| "rewards/GIT": 0.38624581694602966, |
| "rewards/HPSv2": 0.2577781677246094, |
| "rewards/ORM": 0.5109414905309677, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.0, |
| "step": 444 |
| }, |
| { |
| "completion_length": 30.0, |
| "epoch": 0.49280177187153934, |
| "grad_norm": 0.7532635927200317, |
| "kl": 0.041259765625, |
| "learning_rate": 7.21875e-07, |
| "loss": 0.010319485329091549, |
| "reward": 2.7416329383850098, |
| "reward_std": 0.13061758875846863, |
| "rewards/GDino": 0.879687488079071, |
| "rewards/GIT": 0.6991226822137833, |
| "rewards/HPSv2": 0.2878227233886719, |
| "rewards/ORM": 0.875, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.8125, |
| "step": 445 |
| }, |
| { |
| "completion_length": 38.9375, |
| "epoch": 0.4939091915836102, |
| "grad_norm": 0.4900251030921936, |
| "kl": 0.052490234375, |
| "learning_rate": 7.212499999999999e-07, |
| "loss": 0.0012075770646333694, |
| "reward": 2.3121979236602783, |
| "reward_std": 0.12953293696045876, |
| "rewards/GDino": 0.8581786453723907, |
| "rewards/GIT": 0.47472766786813736, |
| "rewards/HPSv2": 0.2680034637451172, |
| "rewards/ORM": 0.711288183927536, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.8125, |
| "step": 446 |
| }, |
| { |
| "completion_length": 43.234375, |
| "epoch": 0.4950166112956811, |
| "grad_norm": 0.5694748759269714, |
| "kl": 0.0531005859375, |
| "learning_rate": 7.206249999999999e-07, |
| "loss": -0.01422290992923081, |
| "reward": 2.2101300954818726, |
| "reward_std": 0.2848183512687683, |
| "rewards/GDino": 0.7993446290493011, |
| "rewards/GIT": 0.4786403179168701, |
| "rewards/HPSv2": 0.27276611328125, |
| "rewards/ORM": 0.6593789905309677, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -23.0625, |
| "step": 447 |
| }, |
| { |
| "completion_length": 40.34375, |
| "epoch": 0.49612403100775193, |
| "grad_norm": 0.5506007671356201, |
| "kl": 0.041748046875, |
| "learning_rate": 7.2e-07, |
| "loss": 0.02106620604172349, |
| "reward": 2.4321142435073853, |
| "reward_std": 0.41992174088954926, |
| "rewards/GDino": 0.8411885499954224, |
| "rewards/GIT": 0.4576234966516495, |
| "rewards/HPSv2": 0.2591552734375, |
| "rewards/ORM": 0.8741468787193298, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.4375, |
| "step": 448 |
| }, |
| { |
| "completion_length": 40.078125, |
| "epoch": 0.49723145071982283, |
| "grad_norm": 0.6929682493209839, |
| "kl": 0.046630859375, |
| "learning_rate": 7.19375e-07, |
| "loss": 0.0026401374489068985, |
| "reward": 2.082669258117676, |
| "reward_std": 0.22932946681976318, |
| "rewards/GDino": 0.8020833730697632, |
| "rewards/GIT": 0.363376647233963, |
| "rewards/HPSv2": 0.28600120544433594, |
| "rewards/ORM": 0.6312080323696136, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.6875, |
| "step": 449 |
| }, |
| { |
| "completion_length": 42.890625, |
| "epoch": 0.4983388704318937, |
| "grad_norm": 0.6465428471565247, |
| "kl": 0.0411376953125, |
| "learning_rate": 7.1875e-07, |
| "loss": -0.003926795674487948, |
| "reward": 1.7539690732955933, |
| "reward_std": 0.381552517414093, |
| "rewards/GDino": 0.8213628232479095, |
| "rewards/GIT": 0.3709906339645386, |
| "rewards/HPSv2": 0.2616615295410156, |
| "rewards/ORM": 0.29995405673980713, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -22.125, |
| "step": 450 |
| }, |
| { |
| "completion_length": 46.078125, |
| "epoch": 0.4994462901439646, |
| "grad_norm": 0.4404943883419037, |
| "kl": 0.0408935546875, |
| "learning_rate": 7.18125e-07, |
| "loss": -0.0011517745442688465, |
| "reward": 1.81854248046875, |
| "reward_std": 0.21245457977056503, |
| "rewards/GDino": 0.7781844735145569, |
| "rewards/GIT": 0.4284133017063141, |
| "rewards/HPSv2": 0.2648448944091797, |
| "rewards/ORM": 0.34709984064102173, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.5625, |
| "step": 451 |
| }, |
| { |
| "completion_length": 48.15625, |
| "epoch": 0.5005537098560354, |
| "grad_norm": 0.7872177958488464, |
| "kl": 0.0487060546875, |
| "learning_rate": 7.175e-07, |
| "loss": -0.008891448844224215, |
| "reward": 2.0443437695503235, |
| "reward_std": 0.3621833920478821, |
| "rewards/GDino": 0.7558982670307159, |
| "rewards/GIT": 0.328241229057312, |
| "rewards/HPSv2": 0.27618408203125, |
| "rewards/ORM": 0.6840202808380127, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.1875, |
| "step": 452 |
| }, |
| { |
| "completion_length": 41.515625, |
| "epoch": 0.5016611295681063, |
| "grad_norm": 0.6438757181167603, |
| "kl": 0.0477294921875, |
| "learning_rate": 7.16875e-07, |
| "loss": 0.013278962578624487, |
| "reward": 2.3407278060913086, |
| "reward_std": 0.29479558765888214, |
| "rewards/GDino": 0.7766183912754059, |
| "rewards/GIT": 0.6374973952770233, |
| "rewards/HPSv2": 0.2586193084716797, |
| "rewards/ORM": 0.6679927408695221, |
| "self_certainty_semantic": -24.75, |
| "self_certainty_token": -22.3125, |
| "step": 453 |
| }, |
| { |
| "completion_length": 44.0625, |
| "epoch": 0.5027685492801772, |
| "grad_norm": 0.609624445438385, |
| "kl": 0.0462646484375, |
| "learning_rate": 7.1625e-07, |
| "loss": -0.0013964123791083694, |
| "reward": 2.5196316242218018, |
| "reward_std": 0.3144952356815338, |
| "rewards/GDino": 0.8671875, |
| "rewards/GIT": 0.6943239569664001, |
| "rewards/HPSv2": 0.2564697265625, |
| "rewards/ORM": 0.7016504108905792, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.3125, |
| "step": 454 |
| }, |
| { |
| "completion_length": 44.859375, |
| "epoch": 0.5038759689922481, |
| "grad_norm": 0.5280178189277649, |
| "kl": 0.04541015625, |
| "learning_rate": 7.156249999999999e-07, |
| "loss": -0.027535758446902037, |
| "reward": 1.9731789231300354, |
| "reward_std": 0.2811855897307396, |
| "rewards/GDino": 0.6670139133930206, |
| "rewards/GIT": 0.3297960162162781, |
| "rewards/HPSv2": 0.2701148986816406, |
| "rewards/ORM": 0.7062540054321289, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.375, |
| "step": 455 |
| }, |
| { |
| "completion_length": 49.84375, |
| "epoch": 0.5049833887043189, |
| "grad_norm": 0.4724181294441223, |
| "kl": 0.041259765625, |
| "learning_rate": 7.149999999999999e-07, |
| "loss": -0.008982280036434531, |
| "reward": 2.0365023612976074, |
| "reward_std": 0.37235787510871887, |
| "rewards/GDino": 0.7172095775604248, |
| "rewards/GIT": 0.42584167420864105, |
| "rewards/HPSv2": 0.2715301513671875, |
| "rewards/ORM": 0.6219209432601929, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -22.1875, |
| "step": 456 |
| }, |
| { |
| "completion_length": 44.765625, |
| "epoch": 0.5060908084163898, |
| "grad_norm": 0.9888372421264648, |
| "kl": 0.0479736328125, |
| "learning_rate": 7.143749999999999e-07, |
| "loss": -0.025254069827497005, |
| "reward": 2.0628309845924377, |
| "reward_std": 0.5208294987678528, |
| "rewards/GDino": 0.7491490542888641, |
| "rewards/GIT": 0.32147393375635147, |
| "rewards/HPSv2": 0.29552268981933594, |
| "rewards/ORM": 0.6966853141784668, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.6875, |
| "step": 457 |
| }, |
| { |
| "completion_length": 54.09375, |
| "epoch": 0.5071982281284607, |
| "grad_norm": 1.0077027082443237, |
| "kl": 0.0501708984375, |
| "learning_rate": 7.137499999999999e-07, |
| "loss": -0.001974734477698803, |
| "reward": 2.337610900402069, |
| "reward_std": 0.31742343306541443, |
| "rewards/GDino": 0.8954645991325378, |
| "rewards/GIT": 0.6387971341609955, |
| "rewards/HPSv2": 0.2842445373535156, |
| "rewards/ORM": 0.5191046893596649, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.125, |
| "step": 458 |
| }, |
| { |
| "completion_length": 33.984375, |
| "epoch": 0.5083056478405316, |
| "grad_norm": 0.6579184532165527, |
| "kl": 0.0506591796875, |
| "learning_rate": 7.13125e-07, |
| "loss": 0.01580179762095213, |
| "reward": 2.240155339241028, |
| "reward_std": 0.23216913640499115, |
| "rewards/GDino": 0.8633578717708588, |
| "rewards/GIT": 0.6162318885326385, |
| "rewards/HPSv2": 0.26489830017089844, |
| "rewards/ORM": 0.4956671893596649, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.125, |
| "step": 459 |
| }, |
| { |
| "completion_length": 53.359375, |
| "epoch": 0.5094130675526024, |
| "grad_norm": 1.016530156135559, |
| "kl": 0.06396484375, |
| "learning_rate": 7.125e-07, |
| "loss": 0.006311747245490551, |
| "reward": 1.799654245376587, |
| "reward_std": 0.45648741722106934, |
| "rewards/GDino": 0.6749999523162842, |
| "rewards/GIT": 0.25362446904182434, |
| "rewards/HPSv2": 0.2591972351074219, |
| "rewards/ORM": 0.6118325591087341, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.0625, |
| "step": 460 |
| }, |
| { |
| "completion_length": 46.015625, |
| "epoch": 0.5105204872646734, |
| "grad_norm": 0.9420766234397888, |
| "kl": 0.035888671875, |
| "learning_rate": 7.11875e-07, |
| "loss": -0.0029141909908503294, |
| "reward": 1.8296540975570679, |
| "reward_std": 0.4803127646446228, |
| "rewards/GDino": 0.6914098262786865, |
| "rewards/GIT": 0.23670585453510284, |
| "rewards/HPSv2": 0.2769889831542969, |
| "rewards/ORM": 0.624549388885498, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.875, |
| "step": 461 |
| }, |
| { |
| "completion_length": 39.1875, |
| "epoch": 0.5116279069767442, |
| "grad_norm": 0.7059751749038696, |
| "kl": 0.048095703125, |
| "learning_rate": 7.1125e-07, |
| "loss": 0.02869965275749564, |
| "reward": 1.7591858506202698, |
| "reward_std": 0.25960099697113037, |
| "rewards/GDino": 0.6792242527008057, |
| "rewards/GIT": 0.39417822659015656, |
| "rewards/HPSv2": 0.24828338623046875, |
| "rewards/ORM": 0.4375, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.4375, |
| "step": 462 |
| }, |
| { |
| "completion_length": 36.21875, |
| "epoch": 0.512735326688815, |
| "grad_norm": 0.9423550963401794, |
| "kl": 0.0567626953125, |
| "learning_rate": 7.106249999999999e-07, |
| "loss": 0.017706943210214376, |
| "reward": 1.816788673400879, |
| "reward_std": 0.3961597681045532, |
| "rewards/GDino": 0.693546324968338, |
| "rewards/GIT": 0.12270250916481018, |
| "rewards/HPSv2": 0.28178977966308594, |
| "rewards/ORM": 0.71875, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.0, |
| "step": 463 |
| }, |
| { |
| "completion_length": 44.1875, |
| "epoch": 0.5138427464008859, |
| "grad_norm": 0.6439041495323181, |
| "kl": 0.04931640625, |
| "learning_rate": 7.1e-07, |
| "loss": 0.02229080768302083, |
| "reward": 1.8793537616729736, |
| "reward_std": 0.32218582928180695, |
| "rewards/GDino": 0.6969353556632996, |
| "rewards/GIT": 0.3158310651779175, |
| "rewards/HPSv2": 0.25373268127441406, |
| "rewards/ORM": 0.6128546893596649, |
| "self_certainty_semantic": -24.8125, |
| "self_certainty_token": -21.5, |
| "step": 464 |
| }, |
| { |
| "completion_length": 43.890625, |
| "epoch": 0.5149501661129569, |
| "grad_norm": 0.5628231167793274, |
| "kl": 0.0523681640625, |
| "learning_rate": 7.09375e-07, |
| "loss": 0.04102162458002567, |
| "reward": 2.2252548933029175, |
| "reward_std": 0.09786996431648731, |
| "rewards/GDino": 0.7161866128444672, |
| "rewards/GIT": 0.34939881414175034, |
| "rewards/HPSv2": 0.2862739562988281, |
| "rewards/ORM": 0.8733955323696136, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.5625, |
| "step": 465 |
| }, |
| { |
| "completion_length": 48.578125, |
| "epoch": 0.5160575858250277, |
| "grad_norm": 0.6555954813957214, |
| "kl": 0.0419921875, |
| "learning_rate": 7.0875e-07, |
| "loss": -0.019896306097507477, |
| "reward": 1.9761043787002563, |
| "reward_std": 0.3829188644886017, |
| "rewards/GDino": 0.7688298523426056, |
| "rewards/GIT": 0.3276664614677429, |
| "rewards/HPSv2": 0.2706336975097656, |
| "rewards/ORM": 0.6089743673801422, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.5, |
| "step": 466 |
| }, |
| { |
| "completion_length": 44.34375, |
| "epoch": 0.5171650055370985, |
| "grad_norm": 0.5814468860626221, |
| "kl": 0.0567626953125, |
| "learning_rate": 7.08125e-07, |
| "loss": -0.001267156796529889, |
| "reward": 2.0606996417045593, |
| "reward_std": 0.3757035583257675, |
| "rewards/GDino": 0.7373589873313904, |
| "rewards/GIT": 0.3534126728773117, |
| "rewards/HPSv2": 0.2546577453613281, |
| "rewards/ORM": 0.7152702808380127, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0625, |
| "step": 467 |
| }, |
| { |
| "completion_length": 37.09375, |
| "epoch": 0.5182724252491694, |
| "grad_norm": 0.7002123594284058, |
| "kl": 0.060302734375, |
| "learning_rate": 7.075e-07, |
| "loss": 0.014457742450758815, |
| "reward": 2.033020555973053, |
| "reward_std": 0.2667345404624939, |
| "rewards/GDino": 0.8099178373813629, |
| "rewards/GIT": 0.3804194927215576, |
| "rewards/HPSv2": 0.2658119201660156, |
| "rewards/ORM": 0.5768712759017944, |
| "self_certainty_semantic": -24.8125, |
| "self_certainty_token": -22.3125, |
| "step": 468 |
| }, |
| { |
| "completion_length": 35.625, |
| "epoch": 0.5193798449612403, |
| "grad_norm": 0.7956823706626892, |
| "kl": 0.0457763671875, |
| "learning_rate": 7.06875e-07, |
| "loss": -0.002700265496969223, |
| "reward": 1.8269972205162048, |
| "reward_std": 0.31584332883358, |
| "rewards/GDino": 0.6803904473781586, |
| "rewards/GIT": 0.19897165894508362, |
| "rewards/HPSv2": 0.263214111328125, |
| "rewards/ORM": 0.6844209432601929, |
| "self_certainty_semantic": -24.875, |
| "self_certainty_token": -21.875, |
| "step": 469 |
| }, |
| { |
| "completion_length": 39.328125, |
| "epoch": 0.5204872646733112, |
| "grad_norm": 1.1874123811721802, |
| "kl": 0.0533447265625, |
| "learning_rate": 7.0625e-07, |
| "loss": 0.013945101760327816, |
| "reward": 1.7942054867744446, |
| "reward_std": 0.504248172044754, |
| "rewards/GDino": 0.6401041746139526, |
| "rewards/GIT": 0.4749217629432678, |
| "rewards/HPSv2": 0.2554292678833008, |
| "rewards/ORM": 0.42375023663043976, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.0, |
| "step": 470 |
| }, |
| { |
| "completion_length": 43.375, |
| "epoch": 0.521594684385382, |
| "grad_norm": 0.7392427325248718, |
| "kl": 0.05810546875, |
| "learning_rate": 7.056249999999999e-07, |
| "loss": -0.0033886469900608063, |
| "reward": 2.60396945476532, |
| "reward_std": 0.2168251946568489, |
| "rewards/GDino": 0.8484375476837158, |
| "rewards/GIT": 0.5769221186637878, |
| "rewards/HPSv2": 0.27235984802246094, |
| "rewards/ORM": 0.90625, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.125, |
| "step": 471 |
| }, |
| { |
| "completion_length": 41.125, |
| "epoch": 0.5227021040974529, |
| "grad_norm": 0.5476402044296265, |
| "kl": 0.05908203125, |
| "learning_rate": 7.049999999999999e-07, |
| "loss": -6.132526323199272e-06, |
| "reward": 1.770015001296997, |
| "reward_std": 0.42023611068725586, |
| "rewards/GDino": 0.6564308404922485, |
| "rewards/GIT": 0.18880262970924377, |
| "rewards/HPSv2": 0.27821922302246094, |
| "rewards/ORM": 0.6465622782707214, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.0, |
| "step": 472 |
| }, |
| { |
| "completion_length": 37.53125, |
| "epoch": 0.5238095238095238, |
| "grad_norm": 0.5314967632293701, |
| "kl": 0.0400390625, |
| "learning_rate": 7.043749999999999e-07, |
| "loss": 0.008513603825122118, |
| "reward": 2.6033971309661865, |
| "reward_std": 0.18285945057868958, |
| "rewards/GDino": 0.7744791507720947, |
| "rewards/GIT": 0.6311581432819366, |
| "rewards/HPSv2": 0.25057220458984375, |
| "rewards/ORM": 0.9471877217292786, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.625, |
| "step": 473 |
| }, |
| { |
| "completion_length": 40.234375, |
| "epoch": 0.5249169435215947, |
| "grad_norm": 0.5836285352706909, |
| "kl": 0.04296875, |
| "learning_rate": 7.037499999999999e-07, |
| "loss": 0.007610062952153385, |
| "reward": 2.063364267349243, |
| "reward_std": 0.23948609083890915, |
| "rewards/GDino": 0.7378768920898438, |
| "rewards/GIT": 0.3602711111307144, |
| "rewards/HPSv2": 0.26682472229003906, |
| "rewards/ORM": 0.6983915567398071, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.25, |
| "step": 474 |
| }, |
| { |
| "completion_length": 40.453125, |
| "epoch": 0.5260243632336655, |
| "grad_norm": 0.6393899917602539, |
| "kl": 0.0455322265625, |
| "learning_rate": 7.031249999999999e-07, |
| "loss": 0.013668630854226649, |
| "reward": 1.6752720475196838, |
| "reward_std": 0.504058450460434, |
| "rewards/GDino": 0.6374790370464325, |
| "rewards/GIT": 0.31694649904966354, |
| "rewards/HPSv2": 0.2520465850830078, |
| "rewards/ORM": 0.4687999337911606, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.3125, |
| "step": 475 |
| }, |
| { |
| "completion_length": 41.8125, |
| "epoch": 0.5271317829457365, |
| "grad_norm": 0.5585540533065796, |
| "kl": 0.05126953125, |
| "learning_rate": 7.024999999999999e-07, |
| "loss": 0.013834381476044655, |
| "reward": 1.9375313520431519, |
| "reward_std": 0.4423487037420273, |
| "rewards/GDino": 0.7547129392623901, |
| "rewards/GIT": 0.10238027572631836, |
| "rewards/HPSv2": 0.2806549072265625, |
| "rewards/ORM": 0.7997831404209137, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.9375, |
| "step": 476 |
| }, |
| { |
| "completion_length": 33.4375, |
| "epoch": 0.5282392026578073, |
| "grad_norm": 0.6089532971382141, |
| "kl": 0.0474853515625, |
| "learning_rate": 7.01875e-07, |
| "loss": -0.0002646814100444317, |
| "reward": 1.9899710416793823, |
| "reward_std": 0.5306534469127655, |
| "rewards/GDino": 0.6599739193916321, |
| "rewards/GIT": 0.2705283910036087, |
| "rewards/HPSv2": 0.2672252655029297, |
| "rewards/ORM": 0.7922433316707611, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.3125, |
| "step": 477 |
| }, |
| { |
| "completion_length": 39.796875, |
| "epoch": 0.5293466223698782, |
| "grad_norm": 0.6699002385139465, |
| "kl": 0.0443115234375, |
| "learning_rate": 7.0125e-07, |
| "loss": 0.007601808290928602, |
| "reward": 2.5651779174804688, |
| "reward_std": 0.430521622300148, |
| "rewards/GDino": 0.8947426080703735, |
| "rewards/GIT": 0.617702305316925, |
| "rewards/HPSv2": 0.2605915069580078, |
| "rewards/ORM": 0.7921415567398071, |
| "self_certainty_semantic": -24.8125, |
| "self_certainty_token": -21.375, |
| "step": 478 |
| }, |
| { |
| "completion_length": 41.6875, |
| "epoch": 0.530454042081949, |
| "grad_norm": 0.5357719659805298, |
| "kl": 0.069580078125, |
| "learning_rate": 7.006250000000001e-07, |
| "loss": -0.017411263193935156, |
| "reward": 2.1919682025909424, |
| "reward_std": 0.34072282910346985, |
| "rewards/GDino": 0.7859375178813934, |
| "rewards/GIT": 0.5476135611534119, |
| "rewards/HPSv2": 0.26539039611816406, |
| "rewards/ORM": 0.5930267721414566, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.875, |
| "step": 479 |
| }, |
| { |
| "completion_length": 44.296875, |
| "epoch": 0.53156146179402, |
| "grad_norm": 0.4808535575866699, |
| "kl": 0.04248046875, |
| "learning_rate": 7e-07, |
| "loss": -0.012577058747410774, |
| "reward": 1.7866081595420837, |
| "reward_std": 0.3765677809715271, |
| "rewards/GDino": 0.5795279145240784, |
| "rewards/GIT": 0.12062875926494598, |
| "rewards/HPSv2": 0.2770805358886719, |
| "rewards/ORM": 0.8093709945678711, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.4375, |
| "step": 480 |
| }, |
| { |
| "completion_length": 42.03125, |
| "epoch": 0.5326688815060908, |
| "grad_norm": 0.6451517343521118, |
| "kl": 0.0438232421875, |
| "learning_rate": 6.99375e-07, |
| "loss": 0.009285415522754192, |
| "reward": 2.0640020966529846, |
| "reward_std": 0.2619232162833214, |
| "rewards/GDino": 0.6719355881214142, |
| "rewards/GIT": 0.23923633992671967, |
| "rewards/HPSv2": 0.2778301239013672, |
| "rewards/ORM": 0.875, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.75, |
| "step": 481 |
| }, |
| { |
| "completion_length": 37.234375, |
| "epoch": 0.5337763012181617, |
| "grad_norm": 1.7300623655319214, |
| "kl": 0.053466796875, |
| "learning_rate": 6.9875e-07, |
| "loss": -0.008384064538404346, |
| "reward": 2.6090811491012573, |
| "reward_std": 0.2747221887111664, |
| "rewards/GDino": 0.854687511920929, |
| "rewards/GIT": 0.7133489549160004, |
| "rewards/HPSv2": 0.2663536071777344, |
| "rewards/ORM": 0.7746912240982056, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.25, |
| "step": 482 |
| }, |
| { |
| "completion_length": 38.3125, |
| "epoch": 0.5348837209302325, |
| "grad_norm": 0.597758948802948, |
| "kl": 0.0496826171875, |
| "learning_rate": 6.98125e-07, |
| "loss": -0.006705485051497817, |
| "reward": 2.23333203792572, |
| "reward_std": 0.232535682618618, |
| "rewards/GDino": 0.8164223432540894, |
| "rewards/GIT": 0.5625110119581223, |
| "rewards/HPSv2": 0.2606487274169922, |
| "rewards/ORM": 0.59375, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.8125, |
| "step": 483 |
| }, |
| { |
| "completion_length": 40.453125, |
| "epoch": 0.5359911406423035, |
| "grad_norm": 0.4974672496318817, |
| "kl": 0.04736328125, |
| "learning_rate": 6.975e-07, |
| "loss": 0.02514065010473132, |
| "reward": 2.259411573410034, |
| "reward_std": 0.37527331709861755, |
| "rewards/GDino": 0.8667246103286743, |
| "rewards/GIT": 0.5178156942129135, |
| "rewards/HPSv2": 0.254425048828125, |
| "rewards/ORM": 0.6204463392496109, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.625, |
| "step": 484 |
| }, |
| { |
| "completion_length": 39.359375, |
| "epoch": 0.5370985603543743, |
| "grad_norm": 0.5150190591812134, |
| "kl": 0.0445556640625, |
| "learning_rate": 6.96875e-07, |
| "loss": -0.016253040172159672, |
| "reward": 2.223387122154236, |
| "reward_std": 0.4909762889146805, |
| "rewards/GDino": 0.8627164661884308, |
| "rewards/GIT": 0.5708933770656586, |
| "rewards/HPSv2": 0.26195716857910156, |
| "rewards/ORM": 0.5278202295303345, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.75, |
| "step": 485 |
| }, |
| { |
| "completion_length": 49.515625, |
| "epoch": 0.5382059800664452, |
| "grad_norm": 0.6220833659172058, |
| "kl": 0.0450439453125, |
| "learning_rate": 6.9625e-07, |
| "loss": -0.014192406553775072, |
| "reward": 1.9840824604034424, |
| "reward_std": 0.2602059319615364, |
| "rewards/GDino": 0.7390625476837158, |
| "rewards/GIT": 0.6071967780590057, |
| "rewards/HPSv2": 0.26282310485839844, |
| "rewards/ORM": 0.375, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.3125, |
| "step": 486 |
| }, |
| { |
| "completion_length": 53.03125, |
| "epoch": 0.5393133997785161, |
| "grad_norm": 1.2206982374191284, |
| "kl": 0.053466796875, |
| "learning_rate": 6.95625e-07, |
| "loss": -0.028607182204723358, |
| "reward": 1.9813396334648132, |
| "reward_std": 0.25878557562828064, |
| "rewards/GDino": 0.7200272381305695, |
| "rewards/GIT": 0.5861987769603729, |
| "rewards/HPSv2": 0.2532386779785156, |
| "rewards/ORM": 0.421875, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.75, |
| "step": 487 |
| }, |
| { |
| "completion_length": 51.328125, |
| "epoch": 0.540420819490587, |
| "grad_norm": 0.9301148056983948, |
| "kl": 0.0439453125, |
| "learning_rate": 6.949999999999999e-07, |
| "loss": -0.02558594848960638, |
| "reward": 1.8908702731132507, |
| "reward_std": 0.1930691972374916, |
| "rewards/GDino": 0.745572954416275, |
| "rewards/GIT": 0.3518243879079819, |
| "rewards/HPSv2": 0.2747688293457031, |
| "rewards/ORM": 0.5187040269374847, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.125, |
| "step": 488 |
| }, |
| { |
| "completion_length": 42.25, |
| "epoch": 0.5415282392026578, |
| "grad_norm": 0.5403162240982056, |
| "kl": 0.05615234375, |
| "learning_rate": 6.943749999999999e-07, |
| "loss": -0.014918024884536862, |
| "reward": 1.8645083904266357, |
| "reward_std": 0.31887492537498474, |
| "rewards/GDino": 0.7039845585823059, |
| "rewards/GIT": 0.31486817449331284, |
| "rewards/HPSv2": 0.2575721740722656, |
| "rewards/ORM": 0.5880835056304932, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.3125, |
| "step": 489 |
| }, |
| { |
| "completion_length": 53.9375, |
| "epoch": 0.5426356589147286, |
| "grad_norm": 0.6085987687110901, |
| "kl": 0.0477294921875, |
| "learning_rate": 6.937499999999999e-07, |
| "loss": 0.026606434024870396, |
| "reward": 1.5878762006759644, |
| "reward_std": 0.39282889664173126, |
| "rewards/GDino": 0.5839112102985382, |
| "rewards/GIT": 0.23308433592319489, |
| "rewards/HPSv2": 0.25993919372558594, |
| "rewards/ORM": 0.5109414830803871, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -22.0, |
| "step": 490 |
| }, |
| { |
| "completion_length": 43.1875, |
| "epoch": 0.5437430786267996, |
| "grad_norm": 0.7043938040733337, |
| "kl": 0.0709228515625, |
| "learning_rate": 6.931249999999999e-07, |
| "loss": -0.0003202613443136215, |
| "reward": 1.4926478266716003, |
| "reward_std": 0.4612387716770172, |
| "rewards/GDino": 0.5785077512264252, |
| "rewards/GIT": 0.12771567702293396, |
| "rewards/HPSv2": 0.272003173828125, |
| "rewards/ORM": 0.5144212394952774, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -22.1875, |
| "step": 491 |
| }, |
| { |
| "completion_length": 39.0625, |
| "epoch": 0.5448504983388704, |
| "grad_norm": 1.6215249300003052, |
| "kl": 0.0604248046875, |
| "learning_rate": 6.924999999999999e-07, |
| "loss": 0.003573375754058361, |
| "reward": 1.6965920329093933, |
| "reward_std": 0.33052287995815277, |
| "rewards/GDino": 0.7088837027549744, |
| "rewards/GIT": 0.2778472602367401, |
| "rewards/HPSv2": 0.27584075927734375, |
| "rewards/ORM": 0.4340203106403351, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.75, |
| "step": 492 |
| }, |
| { |
| "completion_length": 42.953125, |
| "epoch": 0.5459579180509413, |
| "grad_norm": 0.6043370366096497, |
| "kl": 0.0589599609375, |
| "learning_rate": 6.918749999999999e-07, |
| "loss": 0.0050405459478497505, |
| "reward": 2.4023711681365967, |
| "reward_std": 0.3025038540363312, |
| "rewards/GDino": 0.710416704416275, |
| "rewards/GIT": 0.5207570940256119, |
| "rewards/HPSv2": 0.28365135192871094, |
| "rewards/ORM": 0.8875459730625153, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.9375, |
| "step": 493 |
| }, |
| { |
| "completion_length": 43.625, |
| "epoch": 0.5470653377630121, |
| "grad_norm": 0.5131484270095825, |
| "kl": 0.0443115234375, |
| "learning_rate": 6.9125e-07, |
| "loss": 0.0010174026247113943, |
| "reward": 2.6679136753082275, |
| "reward_std": 0.16655711084604263, |
| "rewards/GDino": 0.8572916984558105, |
| "rewards/GIT": 0.7125807106494904, |
| "rewards/HPSv2": 0.26370811462402344, |
| "rewards/ORM": 0.8343330323696136, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.4375, |
| "step": 494 |
| }, |
| { |
| "completion_length": 46.671875, |
| "epoch": 0.5481727574750831, |
| "grad_norm": 0.5884466171264648, |
| "kl": 0.04248046875, |
| "learning_rate": 6.906250000000001e-07, |
| "loss": 0.004825513111427426, |
| "reward": 2.057460904121399, |
| "reward_std": 0.3856964707374573, |
| "rewards/GDino": 0.8639159798622131, |
| "rewards/GIT": 0.6097074598073959, |
| "rewards/HPSv2": 0.25571250915527344, |
| "rewards/ORM": 0.3281249925494194, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.3125, |
| "step": 495 |
| }, |
| { |
| "completion_length": 46.515625, |
| "epoch": 0.5492801771871539, |
| "grad_norm": 0.9797373414039612, |
| "kl": 0.053466796875, |
| "learning_rate": 6.9e-07, |
| "loss": 0.010836861794814467, |
| "reward": 2.183039903640747, |
| "reward_std": 0.3088700920343399, |
| "rewards/GDino": 0.87109375, |
| "rewards/GIT": 0.5162058174610138, |
| "rewards/HPSv2": 0.2704277038574219, |
| "rewards/ORM": 0.5253127366304398, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.25, |
| "step": 496 |
| }, |
| { |
| "completion_length": 42.328125, |
| "epoch": 0.5503875968992248, |
| "grad_norm": 0.5162177085876465, |
| "kl": 0.052001953125, |
| "learning_rate": 6.89375e-07, |
| "loss": 0.012685938272625208, |
| "reward": 2.2096526622772217, |
| "reward_std": 0.38902442157268524, |
| "rewards/GDino": 0.7760285139083862, |
| "rewards/GIT": 0.30912603437900543, |
| "rewards/HPSv2": 0.2770366668701172, |
| "rewards/ORM": 0.8474613428115845, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.0, |
| "step": 497 |
| }, |
| { |
| "completion_length": 49.71875, |
| "epoch": 0.5514950166112956, |
| "grad_norm": 0.5989282727241516, |
| "kl": 0.06103515625, |
| "learning_rate": 6.8875e-07, |
| "loss": 0.01426611002534628, |
| "reward": 2.2355239391326904, |
| "reward_std": 0.5061717927455902, |
| "rewards/GDino": 0.8239583373069763, |
| "rewards/GIT": 0.5775501430034637, |
| "rewards/HPSv2": 0.26257896423339844, |
| "rewards/ORM": 0.5714363753795624, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0625, |
| "step": 498 |
| }, |
| { |
| "completion_length": 41.828125, |
| "epoch": 0.5526024363233666, |
| "grad_norm": 0.8058765530586243, |
| "kl": 0.052490234375, |
| "learning_rate": 6.88125e-07, |
| "loss": 0.00940165319480002, |
| "reward": 2.1786409616470337, |
| "reward_std": 0.4084944427013397, |
| "rewards/GDino": 0.76282998919487, |
| "rewards/GIT": 0.4385463073849678, |
| "rewards/HPSv2": 0.2747611999511719, |
| "rewards/ORM": 0.7025035321712494, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.6875, |
| "step": 499 |
| }, |
| { |
| "completion_length": 48.0, |
| "epoch": 0.5537098560354374, |
| "grad_norm": 0.7646964192390442, |
| "kl": 0.0482177734375, |
| "learning_rate": 6.875e-07, |
| "loss": 0.0026139335241168737, |
| "reward": 2.6670485734939575, |
| "reward_std": 0.15402748435735703, |
| "rewards/GDino": 0.8484375178813934, |
| "rewards/GIT": 0.7150631248950958, |
| "rewards/HPSv2": 0.2910480499267578, |
| "rewards/ORM": 0.8125, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.5, |
| "step": 500 |
| }, |
| { |
| "completion_length": 46.234375, |
| "epoch": 0.5548172757475083, |
| "grad_norm": 0.6809297800064087, |
| "kl": 0.048828125, |
| "learning_rate": 6.86875e-07, |
| "loss": 0.03922370169311762, |
| "reward": 2.008356213569641, |
| "reward_std": 0.3604857325553894, |
| "rewards/GDino": 0.6921875178813934, |
| "rewards/GIT": 0.5941764116287231, |
| "rewards/HPSv2": 0.26070404052734375, |
| "rewards/ORM": 0.461288183927536, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.5625, |
| "step": 501 |
| }, |
| { |
| "completion_length": 47.171875, |
| "epoch": 0.5559246954595792, |
| "grad_norm": 0.849571168422699, |
| "kl": 0.049560546875, |
| "learning_rate": 6.8625e-07, |
| "loss": -0.007896499708294868, |
| "reward": 2.224105715751648, |
| "reward_std": 0.42195719480514526, |
| "rewards/GDino": 0.8002032041549683, |
| "rewards/GIT": 0.5095578879117966, |
| "rewards/HPSv2": 0.2751941680908203, |
| "rewards/ORM": 0.6391504108905792, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.625, |
| "step": 502 |
| }, |
| { |
| "completion_length": 39.40625, |
| "epoch": 0.5570321151716501, |
| "grad_norm": 0.4625159502029419, |
| "kl": 0.0596923828125, |
| "learning_rate": 6.85625e-07, |
| "loss": -0.011805586516857147, |
| "reward": 1.6174015998840332, |
| "reward_std": 0.4419572353363037, |
| "rewards/GDino": 0.6323729455471039, |
| "rewards/GIT": 0.2029045894742012, |
| "rewards/HPSv2": 0.25756263732910156, |
| "rewards/ORM": 0.5245613753795624, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.9375, |
| "step": 503 |
| }, |
| { |
| "completion_length": 36.578125, |
| "epoch": 0.5581395348837209, |
| "grad_norm": 0.6870086193084717, |
| "kl": 0.05029296875, |
| "learning_rate": 6.85e-07, |
| "loss": 0.007269066758453846, |
| "reward": 1.8199999928474426, |
| "reward_std": 0.36270634829998016, |
| "rewards/GDino": 0.6532475501298904, |
| "rewards/GIT": 0.34954261779785156, |
| "rewards/HPSv2": 0.2609977722167969, |
| "rewards/ORM": 0.5562120378017426, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.8125, |
| "step": 504 |
| }, |
| { |
| "completion_length": 42.75, |
| "epoch": 0.5592469545957918, |
| "grad_norm": 0.8303393721580505, |
| "kl": 0.0489501953125, |
| "learning_rate": 6.843749999999999e-07, |
| "loss": -0.017457664478570223, |
| "reward": 1.8187562227249146, |
| "reward_std": 0.5198449939489365, |
| "rewards/GDino": 0.6842133402824402, |
| "rewards/GIT": 0.2273491695523262, |
| "rewards/HPSv2": 0.2782115936279297, |
| "rewards/ORM": 0.6289821267127991, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.9375, |
| "step": 505 |
| }, |
| { |
| "completion_length": 38.328125, |
| "epoch": 0.5603543743078627, |
| "grad_norm": 0.5167096257209778, |
| "kl": 0.0445556640625, |
| "learning_rate": 6.837499999999999e-07, |
| "loss": -0.04000814538449049, |
| "reward": 2.066583573818207, |
| "reward_std": 0.22248844802379608, |
| "rewards/GDino": 0.738374799489975, |
| "rewards/GIT": 0.5225924551486969, |
| "rewards/HPSv2": 0.2765922546386719, |
| "rewards/ORM": 0.529024064540863, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.0625, |
| "step": 506 |
| }, |
| { |
| "completion_length": 36.359375, |
| "epoch": 0.5614617940199336, |
| "grad_norm": 0.6051176190376282, |
| "kl": 0.0595703125, |
| "learning_rate": 6.831249999999999e-07, |
| "loss": -0.009458316955715418, |
| "reward": 1.9375656843185425, |
| "reward_std": 0.3073306977748871, |
| "rewards/GDino": 0.6690048575401306, |
| "rewards/GIT": 0.319953590631485, |
| "rewards/HPSv2": 0.2732524871826172, |
| "rewards/ORM": 0.6753546893596649, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.4375, |
| "step": 507 |
| }, |
| { |
| "completion_length": 41.671875, |
| "epoch": 0.5625692137320044, |
| "grad_norm": 0.6827051043510437, |
| "kl": 0.05029296875, |
| "learning_rate": 6.824999999999999e-07, |
| "loss": -0.017990117892622948, |
| "reward": 2.437025785446167, |
| "reward_std": 0.23294425010681152, |
| "rewards/GDino": 0.8168140351772308, |
| "rewards/GIT": 0.5631050169467926, |
| "rewards/HPSv2": 0.2876014709472656, |
| "rewards/ORM": 0.7695053219795227, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.9375, |
| "step": 508 |
| }, |
| { |
| "completion_length": 45.40625, |
| "epoch": 0.5636766334440753, |
| "grad_norm": 0.5109193325042725, |
| "kl": 0.05078125, |
| "learning_rate": 6.818749999999999e-07, |
| "loss": 0.009160971734672785, |
| "reward": 2.4056625366210938, |
| "reward_std": 0.36375945806503296, |
| "rewards/GDino": 0.8914958536624908, |
| "rewards/GIT": 0.539769321680069, |
| "rewards/HPSv2": 0.2717247009277344, |
| "rewards/ORM": 0.70267254114151, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.1875, |
| "step": 509 |
| }, |
| { |
| "completion_length": 42.5, |
| "epoch": 0.5647840531561462, |
| "grad_norm": 0.5111158490180969, |
| "kl": 0.0447998046875, |
| "learning_rate": 6.8125e-07, |
| "loss": -0.0029866271652281284, |
| "reward": 1.9386223554611206, |
| "reward_std": 0.4446902275085449, |
| "rewards/GDino": 0.6332381367683411, |
| "rewards/GIT": 0.3204441964626312, |
| "rewards/HPSv2": 0.28069114685058594, |
| "rewards/ORM": 0.7042488753795624, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -22.5625, |
| "step": 510 |
| }, |
| { |
| "completion_length": 50.6875, |
| "epoch": 0.5658914728682171, |
| "grad_norm": 1.1105425357818604, |
| "kl": 0.0596923828125, |
| "learning_rate": 6.80625e-07, |
| "loss": 0.012685866095125675, |
| "reward": 1.8053985834121704, |
| "reward_std": 0.3349771797657013, |
| "rewards/GDino": 0.6785425245761871, |
| "rewards/GIT": 0.5215012729167938, |
| "rewards/HPSv2": 0.2517871856689453, |
| "rewards/ORM": 0.35356760025024414, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.0625, |
| "step": 511 |
| }, |
| { |
| "completion_length": 56.921875, |
| "epoch": 0.5669988925802879, |
| "grad_norm": 0.5685505867004395, |
| "kl": 0.0518798828125, |
| "learning_rate": 6.800000000000001e-07, |
| "loss": 0.01260447409003973, |
| "reward": 1.8695366978645325, |
| "reward_std": 0.4191875457763672, |
| "rewards/GDino": 0.6864760220050812, |
| "rewards/GIT": 0.2722553610801697, |
| "rewards/HPSv2": 0.2592887878417969, |
| "rewards/ORM": 0.6515165269374847, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.6875, |
| "step": 512 |
| }, |
| { |
| "completion_length": 41.671875, |
| "epoch": 0.5681063122923588, |
| "grad_norm": 0.7231737971305847, |
| "kl": 0.0457763671875, |
| "learning_rate": 6.79375e-07, |
| "loss": -0.0018274849280714989, |
| "reward": 2.0952327251434326, |
| "reward_std": 0.35603393614292145, |
| "rewards/GDino": 0.7697945237159729, |
| "rewards/GIT": 0.5224983841180801, |
| "rewards/HPSv2": 0.2513313293457031, |
| "rewards/ORM": 0.5516084432601929, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0, |
| "step": 513 |
| }, |
| { |
| "completion_length": 42.625, |
| "epoch": 0.5692137320044297, |
| "grad_norm": 0.5151664614677429, |
| "kl": 0.058837890625, |
| "learning_rate": 6.7875e-07, |
| "loss": 0.012911509722471237, |
| "reward": 1.971530795097351, |
| "reward_std": 0.5368378758430481, |
| "rewards/GDino": 0.6933850646018982, |
| "rewards/GIT": 0.315007820725441, |
| "rewards/HPSv2": 0.263092041015625, |
| "rewards/ORM": 0.7000459432601929, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.625, |
| "step": 514 |
| }, |
| { |
| "completion_length": 49.640625, |
| "epoch": 0.5703211517165006, |
| "grad_norm": 0.6750500202178955, |
| "kl": 0.063720703125, |
| "learning_rate": 6.78125e-07, |
| "loss": -0.012025434523820877, |
| "reward": 1.9398669004440308, |
| "reward_std": 0.3209155201911926, |
| "rewards/GDino": 0.6847161054611206, |
| "rewards/GIT": 0.30489368736743927, |
| "rewards/HPSv2": 0.2687225341796875, |
| "rewards/ORM": 0.6815346777439117, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.5, |
| "step": 515 |
| }, |
| { |
| "completion_length": 41.390625, |
| "epoch": 0.5714285714285714, |
| "grad_norm": 0.6152828931808472, |
| "kl": 0.0460205078125, |
| "learning_rate": 6.775e-07, |
| "loss": 0.009829665534198284, |
| "reward": 2.2633460760116577, |
| "reward_std": 0.21108636260032654, |
| "rewards/GDino": 0.8759043514728546, |
| "rewards/GIT": 0.5751205831766129, |
| "rewards/HPSv2": 0.2716541290283203, |
| "rewards/ORM": 0.5406669825315475, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.9375, |
| "step": 516 |
| }, |
| { |
| "completion_length": 43.453125, |
| "epoch": 0.5725359911406424, |
| "grad_norm": 0.793476402759552, |
| "kl": 0.060546875, |
| "learning_rate": 6.76875e-07, |
| "loss": -0.0033091302029788494, |
| "reward": 1.8790732622146606, |
| "reward_std": 0.2955966591835022, |
| "rewards/GDino": 0.7067020535469055, |
| "rewards/GIT": 0.2911861911416054, |
| "rewards/HPSv2": 0.2660026550292969, |
| "rewards/ORM": 0.6151823997497559, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.625, |
| "step": 517 |
| }, |
| { |
| "completion_length": 54.265625, |
| "epoch": 0.5736434108527132, |
| "grad_norm": 0.8740166425704956, |
| "kl": 0.0469970703125, |
| "learning_rate": 6.7625e-07, |
| "loss": 0.008451959118247032, |
| "reward": 2.325771689414978, |
| "reward_std": 0.2714989632368088, |
| "rewards/GDino": 0.8236979246139526, |
| "rewards/GIT": 0.5366794764995575, |
| "rewards/HPSv2": 0.2731609344482422, |
| "rewards/ORM": 0.6922334432601929, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.5, |
| "step": 518 |
| }, |
| { |
| "completion_length": 43.953125, |
| "epoch": 0.574750830564784, |
| "grad_norm": 0.7303571701049805, |
| "kl": 0.044189453125, |
| "learning_rate": 6.75625e-07, |
| "loss": 0.022691648453474045, |
| "reward": 2.4649245738983154, |
| "reward_std": 0.3035038635134697, |
| "rewards/GDino": 0.7510416805744171, |
| "rewards/GIT": 0.6965360343456268, |
| "rewards/HPSv2": 0.26261329650878906, |
| "rewards/ORM": 0.7547334432601929, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.3125, |
| "step": 519 |
| }, |
| { |
| "completion_length": 41.421875, |
| "epoch": 0.5758582502768549, |
| "grad_norm": 0.8882773518562317, |
| "kl": 0.0509033203125, |
| "learning_rate": 6.75e-07, |
| "loss": -0.015754117514006793, |
| "reward": 2.516376256942749, |
| "reward_std": 0.41609495878219604, |
| "rewards/GDino": 0.8531250059604645, |
| "rewards/GIT": 0.644033282995224, |
| "rewards/HPSv2": 0.26301002502441406, |
| "rewards/ORM": 0.7562080323696136, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.4375, |
| "step": 520 |
| }, |
| { |
| "completion_length": 46.1875, |
| "epoch": 0.5769656699889258, |
| "grad_norm": 0.6410781145095825, |
| "kl": 0.0504150390625, |
| "learning_rate": 6.743749999999999e-07, |
| "loss": 0.035128320567309856, |
| "reward": 2.121917724609375, |
| "reward_std": 0.28992293775081635, |
| "rewards/GDino": 0.800000011920929, |
| "rewards/GIT": 0.6208551824092865, |
| "rewards/HPSv2": 0.25101661682128906, |
| "rewards/ORM": 0.4500459283590317, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.3125, |
| "step": 521 |
| }, |
| { |
| "completion_length": 50.453125, |
| "epoch": 0.5780730897009967, |
| "grad_norm": 0.8646048903465271, |
| "kl": 0.0638427734375, |
| "learning_rate": 6.737499999999999e-07, |
| "loss": 0.029995298013091087, |
| "reward": 1.3939986824989319, |
| "reward_std": 0.40169383585453033, |
| "rewards/GDino": 0.46041668951511383, |
| "rewards/GIT": 0.14088189601898193, |
| "rewards/HPSv2": 0.26492977142333984, |
| "rewards/ORM": 0.5277703106403351, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -22.0, |
| "step": 522 |
| }, |
| { |
| "completion_length": 40.296875, |
| "epoch": 0.5791805094130675, |
| "grad_norm": 0.675642728805542, |
| "kl": 0.0726318359375, |
| "learning_rate": 6.731249999999999e-07, |
| "loss": -0.021141795441508293, |
| "reward": 1.9347034692764282, |
| "reward_std": 0.49275273084640503, |
| "rewards/GDino": 0.6882331371307373, |
| "rewards/GIT": 0.35019098967313766, |
| "rewards/HPSv2": 0.2879371643066406, |
| "rewards/ORM": 0.6083420813083649, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.875, |
| "step": 523 |
| }, |
| { |
| "completion_length": 42.8125, |
| "epoch": 0.5802879291251384, |
| "grad_norm": 0.8465986847877502, |
| "kl": 0.068359375, |
| "learning_rate": 6.724999999999999e-07, |
| "loss": -0.0018713045865297318, |
| "reward": 2.3469094038009644, |
| "reward_std": 0.3619401305913925, |
| "rewards/GDino": 0.8242942988872528, |
| "rewards/GIT": 0.49616317451000214, |
| "rewards/HPSv2": 0.28113555908203125, |
| "rewards/ORM": 0.7453164756298065, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -22.1875, |
| "step": 524 |
| }, |
| { |
| "completion_length": 53.890625, |
| "epoch": 0.5813953488372093, |
| "grad_norm": 0.5991634726524353, |
| "kl": 0.06494140625, |
| "learning_rate": 6.718749999999999e-07, |
| "loss": 0.0026051076129078865, |
| "reward": 1.6451754570007324, |
| "reward_std": 0.4450817406177521, |
| "rewards/GDino": 0.6027501821517944, |
| "rewards/GIT": 0.18362435698509216, |
| "rewards/HPSv2": 0.2547416687011719, |
| "rewards/ORM": 0.6040592342615128, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.9375, |
| "step": 525 |
| }, |
| { |
| "completion_length": 47.328125, |
| "epoch": 0.5825027685492802, |
| "grad_norm": 0.9581777453422546, |
| "kl": 0.072998046875, |
| "learning_rate": 6.7125e-07, |
| "loss": -0.03072733711451292, |
| "reward": 2.296738624572754, |
| "reward_std": 0.3741179406642914, |
| "rewards/GDino": 0.7462376356124878, |
| "rewards/GIT": 0.4357232600450516, |
| "rewards/HPSv2": 0.28830718994140625, |
| "rewards/ORM": 0.8264706134796143, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.8125, |
| "step": 526 |
| }, |
| { |
| "completion_length": 45.140625, |
| "epoch": 0.583610188261351, |
| "grad_norm": 0.5074369311332703, |
| "kl": 0.0526123046875, |
| "learning_rate": 6.70625e-07, |
| "loss": 0.03284440189599991, |
| "reward": 1.8881090879440308, |
| "reward_std": 0.32576023042201996, |
| "rewards/GDino": 0.7507218718528748, |
| "rewards/GIT": 0.3530580550432205, |
| "rewards/HPSv2": 0.25963783264160156, |
| "rewards/ORM": 0.524691253900528, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.875, |
| "step": 527 |
| }, |
| { |
| "completion_length": 46.203125, |
| "epoch": 0.584717607973422, |
| "grad_norm": 0.5117936730384827, |
| "kl": 0.0582275390625, |
| "learning_rate": 6.7e-07, |
| "loss": 0.03193348180502653, |
| "reward": 2.1283347606658936, |
| "reward_std": 0.35501185059547424, |
| "rewards/GDino": 0.739062488079071, |
| "rewards/GIT": 0.5069300383329391, |
| "rewards/HPSv2": 0.2556877136230469, |
| "rewards/ORM": 0.6266543865203857, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.0, |
| "step": 528 |
| }, |
| { |
| "completion_length": 55.734375, |
| "epoch": 0.5858250276854928, |
| "grad_norm": 1.0097836256027222, |
| "kl": 0.0501708984375, |
| "learning_rate": 6.69375e-07, |
| "loss": 0.0001771491952240467, |
| "reward": 2.1930066347122192, |
| "reward_std": 0.2546389251947403, |
| "rewards/GDino": 0.7655479907989502, |
| "rewards/GIT": 0.4856642931699753, |
| "rewards/HPSv2": 0.25183677673339844, |
| "rewards/ORM": 0.6899575591087341, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.4375, |
| "step": 529 |
| }, |
| { |
| "completion_length": 50.078125, |
| "epoch": 0.5869324473975637, |
| "grad_norm": 0.8520405888557434, |
| "kl": 0.064697265625, |
| "learning_rate": 6.6875e-07, |
| "loss": -0.019556816201657057, |
| "reward": 1.9101565480232239, |
| "reward_std": 0.2041948214173317, |
| "rewards/GDino": 0.607291653752327, |
| "rewards/GIT": 0.3278944045305252, |
| "rewards/HPSv2": 0.2765789031982422, |
| "rewards/ORM": 0.6983915418386459, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.5625, |
| "step": 530 |
| }, |
| { |
| "completion_length": 42.984375, |
| "epoch": 0.5880398671096345, |
| "grad_norm": 0.534782350063324, |
| "kl": 0.07373046875, |
| "learning_rate": 6.68125e-07, |
| "loss": 0.00037076231092214584, |
| "reward": 1.7505834698677063, |
| "reward_std": 0.27002371847629547, |
| "rewards/GDino": 0.5543570518493652, |
| "rewards/GIT": 0.2095174342393875, |
| "rewards/HPSv2": 0.2848377227783203, |
| "rewards/ORM": 0.701871246099472, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.25, |
| "step": 531 |
| }, |
| { |
| "completion_length": 37.921875, |
| "epoch": 0.5891472868217055, |
| "grad_norm": 0.8290634155273438, |
| "kl": 0.049072265625, |
| "learning_rate": 6.675e-07, |
| "loss": -0.007146769668906927, |
| "reward": 2.579927086830139, |
| "reward_std": 0.23733402788639069, |
| "rewards/GDino": 0.8890624940395355, |
| "rewards/GIT": 0.6916349530220032, |
| "rewards/HPSv2": 0.26325035095214844, |
| "rewards/ORM": 0.7359794676303864, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.6875, |
| "step": 532 |
| }, |
| { |
| "completion_length": 43.734375, |
| "epoch": 0.5902547065337763, |
| "grad_norm": 0.6242944598197937, |
| "kl": 0.05859375, |
| "learning_rate": 6.66875e-07, |
| "loss": 0.0026651606895029545, |
| "reward": 2.338057041168213, |
| "reward_std": 0.2587292790412903, |
| "rewards/GDino": 0.7275370955467224, |
| "rewards/GIT": 0.42005516588687897, |
| "rewards/HPSv2": 0.2885475158691406, |
| "rewards/ORM": 0.9019171893596649, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.9375, |
| "step": 533 |
| }, |
| { |
| "completion_length": 41.046875, |
| "epoch": 0.5913621262458472, |
| "grad_norm": 0.6595937013626099, |
| "kl": 0.059326171875, |
| "learning_rate": 6.6625e-07, |
| "loss": -0.010991135379299521, |
| "reward": 2.0818028450012207, |
| "reward_std": 0.30971798300743103, |
| "rewards/GDino": 0.784375011920929, |
| "rewards/GIT": 0.4618041664361954, |
| "rewards/HPSv2": 0.24357986450195312, |
| "rewards/ORM": 0.5920437723398209, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.9375, |
| "step": 534 |
| }, |
| { |
| "completion_length": 38.921875, |
| "epoch": 0.592469545957918, |
| "grad_norm": 0.6231927275657654, |
| "kl": 0.0654296875, |
| "learning_rate": 6.65625e-07, |
| "loss": 0.019878892228007317, |
| "reward": 1.9771916270256042, |
| "reward_std": 0.33388233184814453, |
| "rewards/GDino": 0.5947916507720947, |
| "rewards/GIT": 0.45582324266433716, |
| "rewards/HPSv2": 0.27800941467285156, |
| "rewards/ORM": 0.6485673487186432, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.5, |
| "step": 535 |
| }, |
| { |
| "completion_length": 41.0625, |
| "epoch": 0.593576965669989, |
| "grad_norm": 0.6234171986579895, |
| "kl": 0.0654296875, |
| "learning_rate": 6.65e-07, |
| "loss": 0.011152018792927265, |
| "reward": 1.9419553875923157, |
| "reward_std": 0.3548129200935364, |
| "rewards/GDino": 0.7523582577705383, |
| "rewards/GIT": 0.39946797490119934, |
| "rewards/HPSv2": 0.2728996276855469, |
| "rewards/ORM": 0.517229437828064, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.1875, |
| "step": 536 |
| }, |
| { |
| "completion_length": 45.0625, |
| "epoch": 0.5946843853820598, |
| "grad_norm": 1.1540968418121338, |
| "kl": 0.09033203125, |
| "learning_rate": 6.64375e-07, |
| "loss": 0.013285273686051369, |
| "reward": 2.122566819190979, |
| "reward_std": 0.4439195245504379, |
| "rewards/GDino": 0.7579460740089417, |
| "rewards/GIT": 0.5050171613693237, |
| "rewards/HPSv2": 0.2603168487548828, |
| "rewards/ORM": 0.5992866456508636, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.3125, |
| "step": 537 |
| }, |
| { |
| "completion_length": 48.796875, |
| "epoch": 0.5957918050941307, |
| "grad_norm": 1.2856310606002808, |
| "kl": 0.0609130859375, |
| "learning_rate": 6.637499999999999e-07, |
| "loss": 0.005113992607221007, |
| "reward": 1.5898514986038208, |
| "reward_std": 0.4098888337612152, |
| "rewards/GDino": 0.4893845319747925, |
| "rewards/GIT": 0.10379865020513535, |
| "rewards/HPSv2": 0.29241943359375, |
| "rewards/ORM": 0.7042489051818848, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.9375, |
| "step": 538 |
| }, |
| { |
| "completion_length": 49.84375, |
| "epoch": 0.5968992248062015, |
| "grad_norm": 0.49107345938682556, |
| "kl": 0.0557861328125, |
| "learning_rate": 6.631249999999999e-07, |
| "loss": -0.026968090794980526, |
| "reward": 2.1781532764434814, |
| "reward_std": 0.38456064462661743, |
| "rewards/GDino": 0.7330729365348816, |
| "rewards/GIT": 0.5828709006309509, |
| "rewards/HPSv2": 0.28301048278808594, |
| "rewards/ORM": 0.579198956489563, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.0, |
| "step": 539 |
| }, |
| { |
| "completion_length": 54.734375, |
| "epoch": 0.5980066445182725, |
| "grad_norm": 0.5697686672210693, |
| "kl": 0.0615234375, |
| "learning_rate": 6.624999999999999e-07, |
| "loss": -0.012464655563235283, |
| "reward": 1.8495140075683594, |
| "reward_std": 0.17595785856246948, |
| "rewards/GDino": 0.6983504891395569, |
| "rewards/GIT": 0.305370956659317, |
| "rewards/HPSv2": 0.26691627502441406, |
| "rewards/ORM": 0.5788763463497162, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0625, |
| "step": 540 |
| }, |
| { |
| "completion_length": 48.203125, |
| "epoch": 0.5991140642303433, |
| "grad_norm": 0.4882778525352478, |
| "kl": 0.0517578125, |
| "learning_rate": 6.618749999999999e-07, |
| "loss": 0.0061897411942481995, |
| "reward": 1.9610798954963684, |
| "reward_std": 0.23265547305345535, |
| "rewards/GDino": 0.7828344106674194, |
| "rewards/GIT": 0.5123494043946266, |
| "rewards/HPSv2": 0.2717914581298828, |
| "rewards/ORM": 0.3941046893596649, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -23.0, |
| "step": 541 |
| }, |
| { |
| "completion_length": 46.9375, |
| "epoch": 0.6002214839424141, |
| "grad_norm": 0.8239565491676331, |
| "kl": 0.11328125, |
| "learning_rate": 6.6125e-07, |
| "loss": 0.007412359584122896, |
| "reward": 1.7606784105300903, |
| "reward_std": 0.41101209819316864, |
| "rewards/GDino": 0.6593928635120392, |
| "rewards/GIT": 0.24423927813768387, |
| "rewards/HPSv2": 0.2684822082519531, |
| "rewards/ORM": 0.5885640680789948, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.5, |
| "step": 542 |
| }, |
| { |
| "completion_length": 44.96875, |
| "epoch": 0.6013289036544851, |
| "grad_norm": 0.5161306858062744, |
| "kl": 0.053955078125, |
| "learning_rate": 6.60625e-07, |
| "loss": 0.019626203924417496, |
| "reward": 1.916968584060669, |
| "reward_std": 0.350626140832901, |
| "rewards/GDino": 0.8481511175632477, |
| "rewards/GIT": 0.2836497351527214, |
| "rewards/HPSv2": 0.26445865631103516, |
| "rewards/ORM": 0.5207091271877289, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0625, |
| "step": 543 |
| }, |
| { |
| "completion_length": 45.203125, |
| "epoch": 0.602436323366556, |
| "grad_norm": 0.7409946918487549, |
| "kl": 0.0550537109375, |
| "learning_rate": 6.6e-07, |
| "loss": 0.0009375737281516194, |
| "reward": 2.439863920211792, |
| "reward_std": 0.417378693819046, |
| "rewards/GDino": 0.8464202582836151, |
| "rewards/GIT": 0.6017903387546539, |
| "rewards/HPSv2": 0.2494659423828125, |
| "rewards/ORM": 0.7421875, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.375, |
| "step": 544 |
| }, |
| { |
| "completion_length": 41.953125, |
| "epoch": 0.6035437430786268, |
| "grad_norm": 1.7777810096740723, |
| "kl": 0.0738525390625, |
| "learning_rate": 6.59375e-07, |
| "loss": -0.0030171696562319994, |
| "reward": 2.1651780009269714, |
| "reward_std": 0.23080424219369888, |
| "rewards/GDino": 0.7197917103767395, |
| "rewards/GIT": 0.6421350538730621, |
| "rewards/HPSv2": 0.2720012664794922, |
| "rewards/ORM": 0.53125, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.0625, |
| "step": 545 |
| }, |
| { |
| "completion_length": 56.328125, |
| "epoch": 0.6046511627906976, |
| "grad_norm": 0.6303988695144653, |
| "kl": 0.072998046875, |
| "learning_rate": 6.587499999999999e-07, |
| "loss": 0.018343636766076088, |
| "reward": 1.5270656943321228, |
| "reward_std": 0.33213473856449127, |
| "rewards/GDino": 0.6508311331272125, |
| "rewards/GIT": 0.3160218670964241, |
| "rewards/HPSv2": 0.25780296325683594, |
| "rewards/ORM": 0.30240975320339203, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.6875, |
| "step": 546 |
| }, |
| { |
| "completion_length": 47.578125, |
| "epoch": 0.6057585825027686, |
| "grad_norm": 2.654705762863159, |
| "kl": 0.38427734375, |
| "learning_rate": 6.581249999999999e-07, |
| "loss": -0.01044553192332387, |
| "reward": 2.2832573652267456, |
| "reward_std": 0.3163461983203888, |
| "rewards/GDino": 0.7944755554199219, |
| "rewards/GIT": 0.3390696793794632, |
| "rewards/HPSv2": 0.2950706481933594, |
| "rewards/ORM": 0.8546415269374847, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.3125, |
| "step": 547 |
| }, |
| { |
| "completion_length": 46.03125, |
| "epoch": 0.6068660022148394, |
| "grad_norm": 0.6309791803359985, |
| "kl": 0.064697265625, |
| "learning_rate": 6.575e-07, |
| "loss": 0.018591823987662792, |
| "reward": 2.0436050295829773, |
| "reward_std": 0.3530414402484894, |
| "rewards/GDino": 0.6515625417232513, |
| "rewards/GIT": 0.35474079102277756, |
| "rewards/HPSv2": 0.2654685974121094, |
| "rewards/ORM": 0.771833062171936, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.8125, |
| "step": 548 |
| }, |
| { |
| "completion_length": 43.3125, |
| "epoch": 0.6079734219269103, |
| "grad_norm": 1.3774304389953613, |
| "kl": 0.069580078125, |
| "learning_rate": 6.56875e-07, |
| "loss": 0.01293690619058907, |
| "reward": 2.2617204189300537, |
| "reward_std": 0.34871113300323486, |
| "rewards/GDino": 0.7810466587543488, |
| "rewards/GIT": 0.5572087168693542, |
| "rewards/HPSv2": 0.270965576171875, |
| "rewards/ORM": 0.6524995565414429, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.5, |
| "step": 549 |
| }, |
| { |
| "completion_length": 45.65625, |
| "epoch": 0.6090808416389811, |
| "grad_norm": 1.2470241785049438, |
| "kl": 0.0673828125, |
| "learning_rate": 6.5625e-07, |
| "loss": 0.02263116929680109, |
| "reward": 2.3779579401016235, |
| "reward_std": 0.21147555857896805, |
| "rewards/GDino": 0.7546875476837158, |
| "rewards/GIT": 0.45893964171409607, |
| "rewards/HPSv2": 0.26897239685058594, |
| "rewards/ORM": 0.8953584432601929, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.9375, |
| "step": 550 |
| }, |
| { |
| "completion_length": 43.9375, |
| "epoch": 0.6101882613510521, |
| "grad_norm": 0.9305184483528137, |
| "kl": 0.088623046875, |
| "learning_rate": 6.55625e-07, |
| "loss": 0.002473963424563408, |
| "reward": 2.317904472351074, |
| "reward_std": 0.35045669972896576, |
| "rewards/GDino": 0.8548519015312195, |
| "rewards/GIT": 0.43362539261579514, |
| "rewards/HPSv2": 0.29938507080078125, |
| "rewards/ORM": 0.7300421893596649, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -23.1875, |
| "step": 551 |
| }, |
| { |
| "completion_length": 37.625, |
| "epoch": 0.6112956810631229, |
| "grad_norm": 0.545630156993866, |
| "kl": 0.0693359375, |
| "learning_rate": 6.55e-07, |
| "loss": -0.002531451638787985, |
| "reward": 2.08464252948761, |
| "reward_std": 0.323154479265213, |
| "rewards/GDino": 0.8056005537509918, |
| "rewards/GIT": 0.5031750798225403, |
| "rewards/HPSv2": 0.2711334228515625, |
| "rewards/ORM": 0.5047334432601929, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.625, |
| "step": 552 |
| }, |
| { |
| "completion_length": 46.0625, |
| "epoch": 0.6124031007751938, |
| "grad_norm": 0.7531204223632812, |
| "kl": 0.0604248046875, |
| "learning_rate": 6.54375e-07, |
| "loss": -0.017301646526902914, |
| "reward": 1.8435069918632507, |
| "reward_std": 0.4002540707588196, |
| "rewards/GDino": 0.6233262568712234, |
| "rewards/GIT": 0.20414622128009796, |
| "rewards/HPSv2": 0.27692604064941406, |
| "rewards/ORM": 0.7391084432601929, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.625, |
| "step": 553 |
| }, |
| { |
| "completion_length": 49.9375, |
| "epoch": 0.6135105204872646, |
| "grad_norm": 0.5987286567687988, |
| "kl": 0.087158203125, |
| "learning_rate": 6.5375e-07, |
| "loss": -0.010272628860548139, |
| "reward": 1.7656413316726685, |
| "reward_std": 0.474995493888855, |
| "rewards/GDino": 0.6400487124919891, |
| "rewards/GIT": 0.2195979207754135, |
| "rewards/HPSv2": 0.27911949157714844, |
| "rewards/ORM": 0.6268752217292786, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.3125, |
| "step": 554 |
| }, |
| { |
| "completion_length": 49.140625, |
| "epoch": 0.6146179401993356, |
| "grad_norm": 0.5443540215492249, |
| "kl": 0.0611572265625, |
| "learning_rate": 6.531249999999999e-07, |
| "loss": -0.013051381334662437, |
| "reward": 1.713015377521515, |
| "reward_std": 0.4085501432418823, |
| "rewards/GDino": 0.6116544008255005, |
| "rewards/GIT": 0.3198617100715637, |
| "rewards/HPSv2": 0.2673988342285156, |
| "rewards/ORM": 0.5141004621982574, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -22.5625, |
| "step": 555 |
| }, |
| { |
| "completion_length": 45.171875, |
| "epoch": 0.6157253599114064, |
| "grad_norm": 0.6643732786178589, |
| "kl": 0.081298828125, |
| "learning_rate": 6.524999999999999e-07, |
| "loss": -0.016484496649354696, |
| "reward": 2.402197480201721, |
| "reward_std": 0.4949145019054413, |
| "rewards/GDino": 0.8265624642372131, |
| "rewards/GIT": 0.5415551960468292, |
| "rewards/HPSv2": 0.24849700927734375, |
| "rewards/ORM": 0.7855828106403351, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -20.875, |
| "step": 556 |
| }, |
| { |
| "completion_length": 39.078125, |
| "epoch": 0.6168327796234773, |
| "grad_norm": 0.6011128425598145, |
| "kl": 0.06201171875, |
| "learning_rate": 6.51875e-07, |
| "loss": 0.0228618448600173, |
| "reward": 1.9158042073249817, |
| "reward_std": 0.4142046123743057, |
| "rewards/GDino": 0.7750618755817413, |
| "rewards/GIT": 0.25893043726682663, |
| "rewards/HPSv2": 0.27260780334472656, |
| "rewards/ORM": 0.609204113483429, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.6875, |
| "step": 557 |
| }, |
| { |
| "completion_length": 51.59375, |
| "epoch": 0.6179401993355482, |
| "grad_norm": 0.657895565032959, |
| "kl": 0.052978515625, |
| "learning_rate": 6.5125e-07, |
| "loss": -0.04735218919813633, |
| "reward": 1.7417017817497253, |
| "reward_std": 0.2977404296398163, |
| "rewards/GDino": 0.6653882265090942, |
| "rewards/GIT": 0.30595987290143967, |
| "rewards/HPSv2": 0.2765617370605469, |
| "rewards/ORM": 0.49379195272922516, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.125, |
| "step": 558 |
| }, |
| { |
| "completion_length": 44.828125, |
| "epoch": 0.6190476190476191, |
| "grad_norm": 0.6480852365493774, |
| "kl": 0.065673828125, |
| "learning_rate": 6.50625e-07, |
| "loss": 0.022880353964865208, |
| "reward": 2.21392822265625, |
| "reward_std": 0.25253190100193024, |
| "rewards/GDino": 0.7878020107746124, |
| "rewards/GIT": 0.39861495047807693, |
| "rewards/HPSv2": 0.2787151336669922, |
| "rewards/ORM": 0.7487961649894714, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.1875, |
| "step": 559 |
| }, |
| { |
| "completion_length": 43.546875, |
| "epoch": 0.6201550387596899, |
| "grad_norm": 1.5846436023712158, |
| "kl": 0.09521484375, |
| "learning_rate": 6.5e-07, |
| "loss": 0.013549871277064085, |
| "reward": 1.3660652041435242, |
| "reward_std": 0.46586059033870697, |
| "rewards/GDino": 0.5709560066461563, |
| "rewards/GIT": 0.2729204297065735, |
| "rewards/HPSv2": 0.2628517150878906, |
| "rewards/ORM": 0.25933703780174255, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.0625, |
| "step": 560 |
| }, |
| { |
| "completion_length": 47.953125, |
| "epoch": 0.6212624584717608, |
| "grad_norm": 0.6772641539573669, |
| "kl": 0.087890625, |
| "learning_rate": 6.49375e-07, |
| "loss": -0.008351297117769718, |
| "reward": 1.8580278754234314, |
| "reward_std": 0.31344228982925415, |
| "rewards/GDino": 0.6416242718696594, |
| "rewards/GIT": 0.347418375313282, |
| "rewards/HPSv2": 0.27796363830566406, |
| "rewards/ORM": 0.5910216569900513, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.6875, |
| "step": 561 |
| }, |
| { |
| "completion_length": 50.21875, |
| "epoch": 0.6223698781838317, |
| "grad_norm": 0.513234555721283, |
| "kl": 0.07275390625, |
| "learning_rate": 6.4875e-07, |
| "loss": -0.012114224955439568, |
| "reward": 2.127378523349762, |
| "reward_std": 0.2737106531858444, |
| "rewards/GDino": 0.825808048248291, |
| "rewards/GIT": 0.4280584901571274, |
| "rewards/HPSv2": 0.27159881591796875, |
| "rewards/ORM": 0.6019132137298584, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.9375, |
| "step": 562 |
| }, |
| { |
| "completion_length": 45.15625, |
| "epoch": 0.6234772978959026, |
| "grad_norm": 0.96412193775177, |
| "kl": 0.0859375, |
| "learning_rate": 6.481249999999999e-07, |
| "loss": -0.03636417631059885, |
| "reward": 2.0414137840270996, |
| "reward_std": 0.4964071810245514, |
| "rewards/GDino": 0.7853887677192688, |
| "rewards/GIT": 0.43233297765254974, |
| "rewards/HPSv2": 0.26239585876464844, |
| "rewards/ORM": 0.5612961649894714, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -22.375, |
| "step": 563 |
| }, |
| { |
| "completion_length": 65.8125, |
| "epoch": 0.6245847176079734, |
| "grad_norm": 0.5135129690170288, |
| "kl": 0.06005859375, |
| "learning_rate": 6.474999999999999e-07, |
| "loss": -0.001872258260846138, |
| "reward": 2.3118677139282227, |
| "reward_std": 0.3188101053237915, |
| "rewards/GDino": 0.830009251832962, |
| "rewards/GIT": 0.41938507556915283, |
| "rewards/HPSv2": 0.2812232971191406, |
| "rewards/ORM": 0.78125, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.875, |
| "step": 564 |
| }, |
| { |
| "completion_length": 47.171875, |
| "epoch": 0.6256921373200443, |
| "grad_norm": 2.157308340072632, |
| "kl": 0.099365234375, |
| "learning_rate": 6.46875e-07, |
| "loss": 0.008208409883081913, |
| "reward": 1.8709299564361572, |
| "reward_std": 0.30264636874198914, |
| "rewards/GDino": 0.7342717945575714, |
| "rewards/GIT": 0.08047330379486084, |
| "rewards/HPSv2": 0.2749347686767578, |
| "rewards/ORM": 0.78125, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.25, |
| "step": 565 |
| }, |
| { |
| "completion_length": 52.296875, |
| "epoch": 0.6267995570321152, |
| "grad_norm": 1.326660394668579, |
| "kl": 0.085693359375, |
| "learning_rate": 6.4625e-07, |
| "loss": -0.00620048982091248, |
| "reward": 1.7999188899993896, |
| "reward_std": 0.1955185905098915, |
| "rewards/GDino": 0.6784451305866241, |
| "rewards/GIT": 0.3365602195262909, |
| "rewards/HPSv2": 0.2645549774169922, |
| "rewards/ORM": 0.5203584432601929, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -23.5, |
| "step": 566 |
| }, |
| { |
| "completion_length": 51.6875, |
| "epoch": 0.627906976744186, |
| "grad_norm": 1.2671253681182861, |
| "kl": 0.068115234375, |
| "learning_rate": 6.45625e-07, |
| "loss": 0.012142400722950697, |
| "reward": 2.22979599237442, |
| "reward_std": 0.3625608831644058, |
| "rewards/GDino": 0.7203125059604645, |
| "rewards/GIT": 0.5460533201694489, |
| "rewards/HPSv2": 0.2686481475830078, |
| "rewards/ORM": 0.6947819888591766, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.1875, |
| "step": 567 |
| }, |
| { |
| "completion_length": 48.21875, |
| "epoch": 0.6290143964562569, |
| "grad_norm": 0.7895189523696899, |
| "kl": 0.0791015625, |
| "learning_rate": 6.45e-07, |
| "loss": -0.01707265805453062, |
| "reward": 1.8749691247940063, |
| "reward_std": 0.39064884185791016, |
| "rewards/GDino": 0.7415328025817871, |
| "rewards/GIT": 0.42620430886745453, |
| "rewards/HPSv2": 0.2739849090576172, |
| "rewards/ORM": 0.4332471340894699, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -21.5, |
| "step": 568 |
| }, |
| { |
| "completion_length": 44.75, |
| "epoch": 0.6301218161683277, |
| "grad_norm": 0.7421867251396179, |
| "kl": 0.0712890625, |
| "learning_rate": 6.44375e-07, |
| "loss": 0.01910808007232845, |
| "reward": 2.3217188119888306, |
| "reward_std": 0.3109547942876816, |
| "rewards/GDino": 0.7889922857284546, |
| "rewards/GIT": 0.5959710776805878, |
| "rewards/HPSv2": 0.2705860137939453, |
| "rewards/ORM": 0.6661693751811981, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.875, |
| "step": 569 |
| }, |
| { |
| "completion_length": 56.25, |
| "epoch": 0.6312292358803987, |
| "grad_norm": 0.5583733320236206, |
| "kl": 0.0609130859375, |
| "learning_rate": 6.4375e-07, |
| "loss": 0.0022791270166635513, |
| "reward": 1.8600765466690063, |
| "reward_std": 0.34499575197696686, |
| "rewards/GDino": 0.7367187440395355, |
| "rewards/GIT": 0.5528475940227509, |
| "rewards/HPSv2": 0.2670764923095703, |
| "rewards/ORM": 0.30343372374773026, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.0, |
| "step": 570 |
| }, |
| { |
| "completion_length": 46.171875, |
| "epoch": 0.6323366555924695, |
| "grad_norm": 0.45145487785339355, |
| "kl": 0.07373046875, |
| "learning_rate": 6.431249999999999e-07, |
| "loss": 0.021353797987103462, |
| "reward": 2.602142810821533, |
| "reward_std": 0.2599358335137367, |
| "rewards/GDino": 0.9237105846405029, |
| "rewards/GIT": 0.6711409389972687, |
| "rewards/HPSv2": 0.27599525451660156, |
| "rewards/ORM": 0.7312959432601929, |
| "self_certainty_semantic": -25.5625, |
| "self_certainty_token": -22.4375, |
| "step": 571 |
| }, |
| { |
| "completion_length": 54.3125, |
| "epoch": 0.6334440753045404, |
| "grad_norm": 1.0492053031921387, |
| "kl": 0.0517578125, |
| "learning_rate": 6.424999999999999e-07, |
| "loss": 0.012994482181966305, |
| "reward": 2.169149875640869, |
| "reward_std": 0.3728819489479065, |
| "rewards/GDino": 0.6718750447034836, |
| "rewards/GIT": 0.4776107966899872, |
| "rewards/HPSv2": 0.2552928924560547, |
| "rewards/ORM": 0.7643712162971497, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -22.75, |
| "step": 572 |
| }, |
| { |
| "completion_length": 54.234375, |
| "epoch": 0.6345514950166113, |
| "grad_norm": 0.47676345705986023, |
| "kl": 0.068359375, |
| "learning_rate": 6.41875e-07, |
| "loss": 0.009478969499468803, |
| "reward": 2.287571668624878, |
| "reward_std": 0.3681768923997879, |
| "rewards/GDino": 0.8140625059604645, |
| "rewards/GIT": 0.5471410304307938, |
| "rewards/HPSv2": 0.26578521728515625, |
| "rewards/ORM": 0.6605828404426575, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.6875, |
| "step": 573 |
| }, |
| { |
| "completion_length": 42.875, |
| "epoch": 0.6356589147286822, |
| "grad_norm": 0.7573015689849854, |
| "kl": 0.07275390625, |
| "learning_rate": 6.4125e-07, |
| "loss": -0.0035428733099251986, |
| "reward": 2.2345346212387085, |
| "reward_std": 0.26802630722522736, |
| "rewards/GDino": 0.699855774641037, |
| "rewards/GIT": 0.45403826981782913, |
| "rewards/HPSv2": 0.27089691162109375, |
| "rewards/ORM": 0.8097435235977173, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.8125, |
| "step": 574 |
| }, |
| { |
| "completion_length": 38.515625, |
| "epoch": 0.636766334440753, |
| "grad_norm": 1.071760892868042, |
| "kl": 0.071533203125, |
| "learning_rate": 6.40625e-07, |
| "loss": -0.007476763799786568, |
| "reward": 2.5980905294418335, |
| "reward_std": 0.24887971580028534, |
| "rewards/GDino": 0.9643161594867706, |
| "rewards/GIT": 0.7012321054935455, |
| "rewards/HPSv2": 0.2537078857421875, |
| "rewards/ORM": 0.6788344085216522, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -22.375, |
| "step": 575 |
| }, |
| { |
| "completion_length": 56.453125, |
| "epoch": 0.6378737541528239, |
| "grad_norm": 1.1068315505981445, |
| "kl": 0.079833984375, |
| "learning_rate": 6.4e-07, |
| "loss": 0.004310366697609425, |
| "reward": 1.926874816417694, |
| "reward_std": 0.3794712871313095, |
| "rewards/GDino": 0.7066236138343811, |
| "rewards/GIT": 0.3324836865067482, |
| "rewards/HPSv2": 0.27245521545410156, |
| "rewards/ORM": 0.615312248468399, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.5625, |
| "step": 576 |
| }, |
| { |
| "completion_length": 54.5625, |
| "epoch": 0.6389811738648948, |
| "grad_norm": 0.6730782389640808, |
| "kl": 0.09765625, |
| "learning_rate": 6.39375e-07, |
| "loss": -0.009048221400007606, |
| "reward": 1.715060293674469, |
| "reward_std": 0.37638457119464874, |
| "rewards/GDino": 0.6072686910629272, |
| "rewards/GIT": 0.28505346179008484, |
| "rewards/HPSv2": 0.2617626190185547, |
| "rewards/ORM": 0.5609754621982574, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.25, |
| "step": 577 |
| }, |
| { |
| "completion_length": 49.078125, |
| "epoch": 0.6400885935769657, |
| "grad_norm": 0.5849512219429016, |
| "kl": 0.078369140625, |
| "learning_rate": 6.3875e-07, |
| "loss": -0.0033787088468670845, |
| "reward": 2.0598644018173218, |
| "reward_std": 0.3881510943174362, |
| "rewards/GDino": 0.7094090580940247, |
| "rewards/GIT": 0.42222118377685547, |
| "rewards/HPSv2": 0.25662994384765625, |
| "rewards/ORM": 0.6716042459011078, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.375, |
| "step": 578 |
| }, |
| { |
| "completion_length": 49.859375, |
| "epoch": 0.6411960132890365, |
| "grad_norm": 2.933789014816284, |
| "kl": 0.073486328125, |
| "learning_rate": 6.38125e-07, |
| "loss": 0.016461540944874287, |
| "reward": 2.3235161304473877, |
| "reward_std": 0.4396570473909378, |
| "rewards/GDino": 0.7677083313465118, |
| "rewards/GIT": 0.5476280003786087, |
| "rewards/HPSv2": 0.2765331268310547, |
| "rewards/ORM": 0.7316466569900513, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.5, |
| "step": 579 |
| }, |
| { |
| "completion_length": 45.90625, |
| "epoch": 0.6423034330011074, |
| "grad_norm": 0.9947912096977234, |
| "kl": 0.072998046875, |
| "learning_rate": 6.374999999999999e-07, |
| "loss": -0.015327150467783213, |
| "reward": 2.086591362953186, |
| "reward_std": 0.34800001978874207, |
| "rewards/GDino": 0.7345395088195801, |
| "rewards/GIT": 0.40558674931526184, |
| "rewards/HPSv2": 0.2715110778808594, |
| "rewards/ORM": 0.6749540567398071, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.75, |
| "step": 580 |
| }, |
| { |
| "completion_length": 44.46875, |
| "epoch": 0.6434108527131783, |
| "grad_norm": 0.6177259683609009, |
| "kl": 0.086669921875, |
| "learning_rate": 6.368749999999999e-07, |
| "loss": 0.011248688213527203, |
| "reward": 1.933541178703308, |
| "reward_std": 0.35859090089797974, |
| "rewards/GDino": 0.6562248468399048, |
| "rewards/GIT": 0.44320014119148254, |
| "rewards/HPSv2": 0.2723674774169922, |
| "rewards/ORM": 0.561748668551445, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.3125, |
| "step": 581 |
| }, |
| { |
| "completion_length": 38.6875, |
| "epoch": 0.6445182724252492, |
| "grad_norm": 0.5649813413619995, |
| "kl": 0.087890625, |
| "learning_rate": 6.362499999999999e-07, |
| "loss": -0.02157511841505766, |
| "reward": 2.146281599998474, |
| "reward_std": 0.467723548412323, |
| "rewards/GDino": 0.817384660243988, |
| "rewards/GIT": 0.5064477473497391, |
| "rewards/HPSv2": 0.2739696502685547, |
| "rewards/ORM": 0.5484794974327087, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.75, |
| "step": 582 |
| }, |
| { |
| "completion_length": 57.4375, |
| "epoch": 0.64562569213732, |
| "grad_norm": 1.1887096166610718, |
| "kl": 0.071044921875, |
| "learning_rate": 6.35625e-07, |
| "loss": -0.009709077654406428, |
| "reward": 1.8209284543991089, |
| "reward_std": 0.4260648190975189, |
| "rewards/GDino": 0.7038745880126953, |
| "rewards/GIT": 0.388409286737442, |
| "rewards/HPSv2": 0.27118682861328125, |
| "rewards/ORM": 0.4574578106403351, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0625, |
| "step": 583 |
| }, |
| { |
| "completion_length": 55.859375, |
| "epoch": 0.646733111849391, |
| "grad_norm": 1.0990617275238037, |
| "kl": 0.10107421875, |
| "learning_rate": 6.35e-07, |
| "loss": 4.3497420847415924e-05, |
| "reward": 1.8322890996932983, |
| "reward_std": 0.47952717542648315, |
| "rewards/GDino": 0.6861795485019684, |
| "rewards/GIT": 0.322199709713459, |
| "rewards/HPSv2": 0.27663421630859375, |
| "rewards/ORM": 0.547275647521019, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.625, |
| "step": 584 |
| }, |
| { |
| "completion_length": 54.171875, |
| "epoch": 0.6478405315614618, |
| "grad_norm": 0.743022620677948, |
| "kl": 0.099365234375, |
| "learning_rate": 6.34375e-07, |
| "loss": -0.007025235798209906, |
| "reward": 1.7467429041862488, |
| "reward_std": 0.3646506369113922, |
| "rewards/GDino": 0.6056530177593231, |
| "rewards/GIT": 0.2070682942867279, |
| "rewards/HPSv2": 0.2825050354003906, |
| "rewards/ORM": 0.6515165567398071, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.625, |
| "step": 585 |
| }, |
| { |
| "completion_length": 47.171875, |
| "epoch": 0.6489479512735327, |
| "grad_norm": 0.9150655269622803, |
| "kl": 0.0694580078125, |
| "learning_rate": 6.3375e-07, |
| "loss": -0.008931130170822144, |
| "reward": 2.2964425086975098, |
| "reward_std": 0.4491335302591324, |
| "rewards/GDino": 0.8038741052150726, |
| "rewards/GIT": 0.5696861445903778, |
| "rewards/HPSv2": 0.2661018371582031, |
| "rewards/ORM": 0.656780481338501, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.875, |
| "step": 586 |
| }, |
| { |
| "completion_length": 48.75, |
| "epoch": 0.6500553709856035, |
| "grad_norm": 0.4439801871776581, |
| "kl": 0.071533203125, |
| "learning_rate": 6.33125e-07, |
| "loss": 0.03945009782910347, |
| "reward": 2.233885407447815, |
| "reward_std": 0.44437268376350403, |
| "rewards/GDino": 0.8034193813800812, |
| "rewards/GIT": 0.47745344042778015, |
| "rewards/HPSv2": 0.26355743408203125, |
| "rewards/ORM": 0.6894551515579224, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.25, |
| "step": 587 |
| }, |
| { |
| "completion_length": 52.828125, |
| "epoch": 0.6511627906976745, |
| "grad_norm": 0.7273818254470825, |
| "kl": 0.104736328125, |
| "learning_rate": 6.324999999999999e-07, |
| "loss": -0.04699191078543663, |
| "reward": 2.418140411376953, |
| "reward_std": 0.25689420104026794, |
| "rewards/GDino": 0.729385107755661, |
| "rewards/GIT": 0.5350233688950539, |
| "rewards/HPSv2": 0.2727947235107422, |
| "rewards/ORM": 0.8809372782707214, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.8125, |
| "step": 588 |
| }, |
| { |
| "completion_length": 52.671875, |
| "epoch": 0.6522702104097453, |
| "grad_norm": 1.8505158424377441, |
| "kl": 0.0869140625, |
| "learning_rate": 6.31875e-07, |
| "loss": -0.020862075500190258, |
| "reward": 2.3336164951324463, |
| "reward_std": 0.19946596771478653, |
| "rewards/GDino": 0.7767625749111176, |
| "rewards/GIT": 0.5896276831626892, |
| "rewards/HPSv2": 0.2785224914550781, |
| "rewards/ORM": 0.6887038052082062, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.9375, |
| "step": 589 |
| }, |
| { |
| "completion_length": 53.5, |
| "epoch": 0.6533776301218162, |
| "grad_norm": 0.5300537347793579, |
| "kl": 0.094482421875, |
| "learning_rate": 6.3125e-07, |
| "loss": -0.010553581640124321, |
| "reward": 2.2597015500068665, |
| "reward_std": 0.3002111464738846, |
| "rewards/GDino": 0.7936822772026062, |
| "rewards/GIT": 0.44989803433418274, |
| "rewards/HPSv2": 0.27987098693847656, |
| "rewards/ORM": 0.7362502217292786, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.9375, |
| "step": 590 |
| }, |
| { |
| "completion_length": 42.5625, |
| "epoch": 0.654485049833887, |
| "grad_norm": 0.583197295665741, |
| "kl": 0.0830078125, |
| "learning_rate": 6.30625e-07, |
| "loss": 0.015054948627948761, |
| "reward": 2.329821825027466, |
| "reward_std": 0.23674922436475754, |
| "rewards/GDino": 0.7871354520320892, |
| "rewards/GIT": 0.6012963056564331, |
| "rewards/HPSv2": 0.2722434997558594, |
| "rewards/ORM": 0.6691466569900513, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.3125, |
| "step": 591 |
| }, |
| { |
| "completion_length": 50.15625, |
| "epoch": 0.655592469545958, |
| "grad_norm": 1.685902714729309, |
| "kl": 0.08154296875, |
| "learning_rate": 6.3e-07, |
| "loss": -0.013527771923691034, |
| "reward": 1.8206582069396973, |
| "reward_std": 0.36716659367084503, |
| "rewards/GDino": 0.675000011920929, |
| "rewards/GIT": 0.4574429839849472, |
| "rewards/HPSv2": 0.2741527557373047, |
| "rewards/ORM": 0.4140624850988388, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.75, |
| "step": 592 |
| }, |
| { |
| "completion_length": 56.390625, |
| "epoch": 0.6566998892580288, |
| "grad_norm": 0.5346245169639587, |
| "kl": 0.067626953125, |
| "learning_rate": 6.29375e-07, |
| "loss": -0.028868052177131176, |
| "reward": 2.3374955654144287, |
| "reward_std": 0.448678582906723, |
| "rewards/GDino": 0.875, |
| "rewards/GIT": 0.5937578082084656, |
| "rewards/HPSv2": 0.2687797546386719, |
| "rewards/ORM": 0.599958062171936, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0625, |
| "step": 593 |
| }, |
| { |
| "completion_length": 53.46875, |
| "epoch": 0.6578073089700996, |
| "grad_norm": 0.5602062940597534, |
| "kl": 0.09130859375, |
| "learning_rate": 6.2875e-07, |
| "loss": 0.014588311780244112, |
| "reward": 2.076681613922119, |
| "reward_std": 0.2987568974494934, |
| "rewards/GDino": 0.7875000238418579, |
| "rewards/GIT": 0.46318748593330383, |
| "rewards/HPSv2": 0.27095603942871094, |
| "rewards/ORM": 0.555038183927536, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.1875, |
| "step": 594 |
| }, |
| { |
| "completion_length": 50.234375, |
| "epoch": 0.6589147286821705, |
| "grad_norm": 0.8814988136291504, |
| "kl": 0.115234375, |
| "learning_rate": 6.28125e-07, |
| "loss": -0.004488900303840637, |
| "reward": 1.7216359376907349, |
| "reward_std": 0.32969560474157333, |
| "rewards/GDino": 0.6518171429634094, |
| "rewards/GIT": 0.2219809964299202, |
| "rewards/HPSv2": 0.2915458679199219, |
| "rewards/ORM": 0.5562919676303864, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0, |
| "step": 595 |
| }, |
| { |
| "completion_length": 44.859375, |
| "epoch": 0.6600221483942414, |
| "grad_norm": 0.6903232336044312, |
| "kl": 0.080322265625, |
| "learning_rate": 6.274999999999999e-07, |
| "loss": -0.0038359593600034714, |
| "reward": 2.2345484495162964, |
| "reward_std": 0.35491831600666046, |
| "rewards/GDino": 0.7885048389434814, |
| "rewards/GIT": 0.49162155389785767, |
| "rewards/HPSv2": 0.2669219970703125, |
| "rewards/ORM": 0.6875, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.625, |
| "step": 596 |
| }, |
| { |
| "completion_length": 44.109375, |
| "epoch": 0.6611295681063123, |
| "grad_norm": 0.6238202452659607, |
| "kl": 0.11376953125, |
| "learning_rate": 6.268749999999999e-07, |
| "loss": 0.010757120093330741, |
| "reward": 1.9933987259864807, |
| "reward_std": 0.43311460316181183, |
| "rewards/GDino": 0.7434804737567902, |
| "rewards/GIT": 0.22067713737487793, |
| "rewards/HPSv2": 0.28347206115722656, |
| "rewards/ORM": 0.7457689642906189, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.5, |
| "step": 597 |
| }, |
| { |
| "completion_length": 47.0, |
| "epoch": 0.6622369878183831, |
| "grad_norm": 0.6406247615814209, |
| "kl": 0.081787109375, |
| "learning_rate": 6.262499999999999e-07, |
| "loss": -0.0002176426351070404, |
| "reward": 2.7149163484573364, |
| "reward_std": 0.18262185156345367, |
| "rewards/GDino": 0.8359375, |
| "rewards/GIT": 0.7479822635650635, |
| "rewards/HPSv2": 0.2559967041015625, |
| "rewards/ORM": 0.875, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.4375, |
| "step": 598 |
| }, |
| { |
| "completion_length": 54.984375, |
| "epoch": 0.6633444075304541, |
| "grad_norm": 0.46733200550079346, |
| "kl": 0.076416015625, |
| "learning_rate": 6.256249999999999e-07, |
| "loss": 0.008587295189499855, |
| "reward": 2.046528160572052, |
| "reward_std": 0.30855168402194977, |
| "rewards/GDino": 0.7562500238418579, |
| "rewards/GIT": 0.4039812982082367, |
| "rewards/HPSv2": 0.26134681701660156, |
| "rewards/ORM": 0.624950036406517, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.4375, |
| "step": 599 |
| }, |
| { |
| "completion_length": 67.96875, |
| "epoch": 0.6644518272425249, |
| "grad_norm": 0.49819114804267883, |
| "kl": 0.067626953125, |
| "learning_rate": 6.249999999999999e-07, |
| "loss": 0.037328097969293594, |
| "reward": 2.071701407432556, |
| "reward_std": 0.4433322548866272, |
| "rewards/GDino": 0.7885022759437561, |
| "rewards/GIT": 0.33964233100414276, |
| "rewards/HPSv2": 0.27628517150878906, |
| "rewards/ORM": 0.6672714054584503, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.4375, |
| "step": 600 |
| }, |
| { |
| "completion_length": 44.9375, |
| "epoch": 0.6655592469545958, |
| "grad_norm": 0.46717965602874756, |
| "kl": 0.1103515625, |
| "learning_rate": 6.24375e-07, |
| "loss": -0.00821652077138424, |
| "reward": 2.10355544090271, |
| "reward_std": 0.40635716915130615, |
| "rewards/GDino": 0.8286458253860474, |
| "rewards/GIT": 0.4935857057571411, |
| "rewards/HPSv2": 0.2825775146484375, |
| "rewards/ORM": 0.49874623119831085, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.25, |
| "step": 601 |
| }, |
| { |
| "completion_length": 57.125, |
| "epoch": 0.6666666666666666, |
| "grad_norm": 0.6258370876312256, |
| "kl": 0.07275390625, |
| "learning_rate": 6.2375e-07, |
| "loss": -0.018634040374308825, |
| "reward": 2.238632559776306, |
| "reward_std": 0.2751503735780716, |
| "rewards/GDino": 0.7313859760761261, |
| "rewards/GIT": 0.3301559388637543, |
| "rewards/HPSv2": 0.28325653076171875, |
| "rewards/ORM": 0.8938339650630951, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -22.4375, |
| "step": 602 |
| }, |
| { |
| "completion_length": 48.078125, |
| "epoch": 0.6677740863787376, |
| "grad_norm": 0.6003196835517883, |
| "kl": 0.10400390625, |
| "learning_rate": 6.23125e-07, |
| "loss": -0.01071217842400074, |
| "reward": 2.2572548389434814, |
| "reward_std": 0.45507198572158813, |
| "rewards/GDino": 0.7749999761581421, |
| "rewards/GIT": 0.5677377134561539, |
| "rewards/HPSv2": 0.25701332092285156, |
| "rewards/ORM": 0.657503753900528, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.25, |
| "step": 603 |
| }, |
| { |
| "completion_length": 56.296875, |
| "epoch": 0.6688815060908084, |
| "grad_norm": 0.9325379729270935, |
| "kl": 0.081298828125, |
| "learning_rate": 6.225000000000001e-07, |
| "loss": 0.0008968440815806389, |
| "reward": 2.599682927131653, |
| "reward_std": 0.31057311594486237, |
| "rewards/GDino": 0.8854166567325592, |
| "rewards/GIT": 0.7147887349128723, |
| "rewards/HPSv2": 0.24934768676757812, |
| "rewards/ORM": 0.7501298785209656, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.4375, |
| "step": 604 |
| }, |
| { |
| "completion_length": 43.859375, |
| "epoch": 0.6699889258028793, |
| "grad_norm": 0.5234432220458984, |
| "kl": 0.08935546875, |
| "learning_rate": 6.21875e-07, |
| "loss": -0.015889065340161324, |
| "reward": 2.258097290992737, |
| "reward_std": 0.26281405985355377, |
| "rewards/GDino": 0.7695927023887634, |
| "rewards/GIT": 0.56023770570755, |
| "rewards/HPSv2": 0.28045082092285156, |
| "rewards/ORM": 0.6478160321712494, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.1875, |
| "step": 605 |
| }, |
| { |
| "completion_length": 51.546875, |
| "epoch": 0.6710963455149501, |
| "grad_norm": 0.6027458906173706, |
| "kl": 0.101806640625, |
| "learning_rate": 6.2125e-07, |
| "loss": -0.000653423834592104, |
| "reward": 1.9265617728233337, |
| "reward_std": 0.41129185259342194, |
| "rewards/GDino": 0.7817370295524597, |
| "rewards/GIT": 0.4215117394924164, |
| "rewards/HPSv2": 0.2592964172363281, |
| "rewards/ORM": 0.46401655673980713, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.1875, |
| "step": 606 |
| }, |
| { |
| "completion_length": 57.421875, |
| "epoch": 0.6722037652270211, |
| "grad_norm": 0.5389038324356079, |
| "kl": 0.077392578125, |
| "learning_rate": 6.20625e-07, |
| "loss": 0.00019401684403419495, |
| "reward": 2.517163872718811, |
| "reward_std": 0.41141896694898605, |
| "rewards/GDino": 0.8523252308368683, |
| "rewards/GIT": 0.7845209240913391, |
| "rewards/HPSv2": 0.2706718444824219, |
| "rewards/ORM": 0.6096457540988922, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.9375, |
| "step": 607 |
| }, |
| { |
| "completion_length": 47.296875, |
| "epoch": 0.6733111849390919, |
| "grad_norm": 0.7910009026527405, |
| "kl": 0.08203125, |
| "learning_rate": 6.2e-07, |
| "loss": -0.02017150539904833, |
| "reward": 1.991845190525055, |
| "reward_std": 0.4512777328491211, |
| "rewards/GDino": 0.7250000238418579, |
| "rewards/GIT": 0.29137129336595535, |
| "rewards/HPSv2": 0.2793083190917969, |
| "rewards/ORM": 0.6961656212806702, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.6875, |
| "step": 608 |
| }, |
| { |
| "completion_length": 49.671875, |
| "epoch": 0.6744186046511628, |
| "grad_norm": 0.6332114934921265, |
| "kl": 0.098388671875, |
| "learning_rate": 6.19375e-07, |
| "loss": 0.0044885119423270226, |
| "reward": 2.611580729484558, |
| "reward_std": 0.48633773624897003, |
| "rewards/GDino": 0.878125011920929, |
| "rewards/GIT": 0.7029311954975128, |
| "rewards/HPSv2": 0.28400421142578125, |
| "rewards/ORM": 0.7465203106403351, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.75, |
| "step": 609 |
| }, |
| { |
| "completion_length": 55.203125, |
| "epoch": 0.6755260243632336, |
| "grad_norm": 0.5662334561347961, |
| "kl": 0.082275390625, |
| "learning_rate": 6.1875e-07, |
| "loss": -0.023096920922398567, |
| "reward": 1.790070652961731, |
| "reward_std": 0.4277941882610321, |
| "rewards/GDino": 0.6870588064193726, |
| "rewards/GIT": 0.23581942915916443, |
| "rewards/HPSv2": 0.26169776916503906, |
| "rewards/ORM": 0.6054946780204773, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.9375, |
| "step": 610 |
| }, |
| { |
| "completion_length": 51.0625, |
| "epoch": 0.6766334440753046, |
| "grad_norm": 0.5869778394699097, |
| "kl": 0.08935546875, |
| "learning_rate": 6.18125e-07, |
| "loss": 0.006669576745480299, |
| "reward": 2.1600626707077026, |
| "reward_std": 0.46393774449825287, |
| "rewards/GDino": 0.7281681597232819, |
| "rewards/GIT": 0.3008827492594719, |
| "rewards/HPSv2": 0.2813243865966797, |
| "rewards/ORM": 0.8496872782707214, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.5, |
| "step": 611 |
| }, |
| { |
| "completion_length": 67.8125, |
| "epoch": 0.6777408637873754, |
| "grad_norm": 0.5921112298965454, |
| "kl": 0.111328125, |
| "learning_rate": 6.175e-07, |
| "loss": -0.014026800054125488, |
| "reward": 2.1289449334144592, |
| "reward_std": 0.3110615909099579, |
| "rewards/GDino": 0.750481367111206, |
| "rewards/GIT": 0.5314053893089294, |
| "rewards/HPSv2": 0.281829833984375, |
| "rewards/ORM": 0.5652283579111099, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.6875, |
| "step": 612 |
| }, |
| { |
| "completion_length": 61.015625, |
| "epoch": 0.6788482834994463, |
| "grad_norm": 1.6785780191421509, |
| "kl": 0.063720703125, |
| "learning_rate": 6.168749999999999e-07, |
| "loss": 0.0001376904547214508, |
| "reward": 2.5762332677841187, |
| "reward_std": 0.1488320305943489, |
| "rewards/GDino": 0.9367609024047852, |
| "rewards/GIT": 0.7700669467449188, |
| "rewards/HPSv2": 0.26944732666015625, |
| "rewards/ORM": 0.5999580472707748, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.25, |
| "step": 613 |
| }, |
| { |
| "completion_length": 61.0, |
| "epoch": 0.6799557032115172, |
| "grad_norm": 0.5101741552352905, |
| "kl": 0.07568359375, |
| "learning_rate": 6.162499999999999e-07, |
| "loss": -0.02052605990320444, |
| "reward": 2.268720507621765, |
| "reward_std": 0.38759973645210266, |
| "rewards/GDino": 0.784375011920929, |
| "rewards/GIT": 0.6036352515220642, |
| "rewards/HPSv2": 0.26512718200683594, |
| "rewards/ORM": 0.6155830025672913, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.5, |
| "step": 614 |
| }, |
| { |
| "completion_length": 45.953125, |
| "epoch": 0.6810631229235881, |
| "grad_norm": 0.7281429171562195, |
| "kl": 0.098876953125, |
| "learning_rate": 6.156249999999999e-07, |
| "loss": 0.007133266655728221, |
| "reward": 1.958534836769104, |
| "reward_std": 0.31903617084026337, |
| "rewards/GDino": 0.8034560978412628, |
| "rewards/GIT": 0.31245744973421097, |
| "rewards/HPSv2": 0.27966880798339844, |
| "rewards/ORM": 0.5629524886608124, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -22.5, |
| "step": 615 |
| }, |
| { |
| "completion_length": 50.5625, |
| "epoch": 0.6821705426356589, |
| "grad_norm": 0.5768497586250305, |
| "kl": 0.094482421875, |
| "learning_rate": 6.149999999999999e-07, |
| "loss": -0.010629929834976792, |
| "reward": 1.4815038442611694, |
| "reward_std": 0.5085962116718292, |
| "rewards/GDino": 0.4787500500679016, |
| "rewards/GIT": 0.18398787081241608, |
| "rewards/HPSv2": 0.24907493591308594, |
| "rewards/ORM": 0.5696910172700882, |
| "self_certainty_semantic": -24.9375, |
| "self_certainty_token": -22.0625, |
| "step": 616 |
| }, |
| { |
| "completion_length": 60.0, |
| "epoch": 0.6832779623477298, |
| "grad_norm": 0.5207316875457764, |
| "kl": 0.086181640625, |
| "learning_rate": 6.143749999999999e-07, |
| "loss": -0.029840716160833836, |
| "reward": 2.400840163230896, |
| "reward_std": 0.25419779121875763, |
| "rewards/GDino": 0.7671903371810913, |
| "rewards/GIT": 0.46047309041023254, |
| "rewards/HPSv2": 0.27982330322265625, |
| "rewards/ORM": 0.8933533728122711, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.625, |
| "step": 617 |
| }, |
| { |
| "completion_length": 54.09375, |
| "epoch": 0.6843853820598007, |
| "grad_norm": 2.1334187984466553, |
| "kl": 0.0654296875, |
| "learning_rate": 6.1375e-07, |
| "loss": -0.012758204014971852, |
| "reward": 2.149357318878174, |
| "reward_std": 0.2887853980064392, |
| "rewards/GDino": 0.7757812738418579, |
| "rewards/GIT": 0.5285128504037857, |
| "rewards/HPSv2": 0.2622547149658203, |
| "rewards/ORM": 0.5828085094690323, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -22.375, |
| "step": 618 |
| }, |
| { |
| "completion_length": 58.890625, |
| "epoch": 0.6854928017718716, |
| "grad_norm": 0.7336542010307312, |
| "kl": 0.118896484375, |
| "learning_rate": 6.13125e-07, |
| "loss": -0.01781146228313446, |
| "reward": 2.0267523527145386, |
| "reward_std": 0.44900983572006226, |
| "rewards/GDino": 0.7080606520175934, |
| "rewards/GIT": 0.3956197053194046, |
| "rewards/HPSv2": 0.2847747802734375, |
| "rewards/ORM": 0.6382972598075867, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -23.0, |
| "step": 619 |
| }, |
| { |
| "completion_length": 60.5625, |
| "epoch": 0.6866002214839424, |
| "grad_norm": 0.8615192174911499, |
| "kl": 0.091064453125, |
| "learning_rate": 6.125000000000001e-07, |
| "loss": 0.013798002619296312, |
| "reward": 2.534471869468689, |
| "reward_std": 0.2539004608988762, |
| "rewards/GDino": 0.8903107047080994, |
| "rewards/GIT": 0.6578922718763351, |
| "rewards/HPSv2": 0.2694740295410156, |
| "rewards/ORM": 0.7167948484420776, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.1875, |
| "step": 620 |
| }, |
| { |
| "completion_length": 48.453125, |
| "epoch": 0.6877076411960132, |
| "grad_norm": 1.2892807722091675, |
| "kl": 0.111328125, |
| "learning_rate": 6.11875e-07, |
| "loss": -0.005828765220940113, |
| "reward": 1.8412444591522217, |
| "reward_std": 0.3962418884038925, |
| "rewards/GDino": 0.6466913968324661, |
| "rewards/GIT": 0.2277315855026245, |
| "rewards/HPSv2": 0.28833770751953125, |
| "rewards/ORM": 0.6784836649894714, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0, |
| "step": 621 |
| }, |
| { |
| "completion_length": 54.515625, |
| "epoch": 0.6888150609080842, |
| "grad_norm": 1.4823932647705078, |
| "kl": 0.124267578125, |
| "learning_rate": 6.1125e-07, |
| "loss": 0.027432953007519245, |
| "reward": 2.020459830760956, |
| "reward_std": 0.4132416397333145, |
| "rewards/GDino": 0.7562499642372131, |
| "rewards/GIT": 0.42041274905204773, |
| "rewards/HPSv2": 0.2812061309814453, |
| "rewards/ORM": 0.5625909566879272, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.3125, |
| "step": 622 |
| }, |
| { |
| "completion_length": 62.0, |
| "epoch": 0.689922480620155, |
| "grad_norm": 0.5628201961517334, |
| "kl": 0.07373046875, |
| "learning_rate": 6.10625e-07, |
| "loss": 0.015298215672373772, |
| "reward": 2.527510166168213, |
| "reward_std": 0.299762986600399, |
| "rewards/GDino": 0.884425014257431, |
| "rewards/GIT": 0.7807052731513977, |
| "rewards/HPSv2": 0.2733135223388672, |
| "rewards/ORM": 0.5890664458274841, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.5625, |
| "step": 623 |
| }, |
| { |
| "completion_length": 62.359375, |
| "epoch": 0.6910299003322259, |
| "grad_norm": 0.7030363082885742, |
| "kl": 0.080810546875, |
| "learning_rate": 6.1e-07, |
| "loss": 0.0059433141723275185, |
| "reward": 1.8672431707382202, |
| "reward_std": 0.36278337240219116, |
| "rewards/GDino": 0.6254703998565674, |
| "rewards/GIT": 0.3477711081504822, |
| "rewards/HPSv2": 0.26592254638671875, |
| "rewards/ORM": 0.6280790567398071, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.1875, |
| "step": 624 |
| }, |
| { |
| "completion_length": 53.890625, |
| "epoch": 0.6921373200442967, |
| "grad_norm": 0.5560760498046875, |
| "kl": 0.085205078125, |
| "learning_rate": 6.09375e-07, |
| "loss": -0.009751995792612433, |
| "reward": 2.183237671852112, |
| "reward_std": 0.48196063935756683, |
| "rewards/GDino": 0.7296874523162842, |
| "rewards/GIT": 0.5111644864082336, |
| "rewards/HPSv2": 0.2658271789550781, |
| "rewards/ORM": 0.6765585243701935, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -21.8125, |
| "step": 625 |
| }, |
| { |
| "completion_length": 52.125, |
| "epoch": 0.6932447397563677, |
| "grad_norm": 0.5723803639411926, |
| "kl": 0.093994140625, |
| "learning_rate": 6.0875e-07, |
| "loss": 0.012737608747556806, |
| "reward": 1.6010103225708008, |
| "reward_std": 0.32132093608379364, |
| "rewards/GDino": 0.7208738923072815, |
| "rewards/GIT": 0.2184068188071251, |
| "rewards/HPSv2": 0.26342201232910156, |
| "rewards/ORM": 0.3983076363801956, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.9375, |
| "step": 626 |
| }, |
| { |
| "completion_length": 60.703125, |
| "epoch": 0.6943521594684385, |
| "grad_norm": 0.541669487953186, |
| "kl": 0.07861328125, |
| "learning_rate": 6.08125e-07, |
| "loss": -0.004021515836939216, |
| "reward": 1.816047489643097, |
| "reward_std": 0.4408939480781555, |
| "rewards/GDino": 0.6644719839096069, |
| "rewards/GIT": 0.3648825138807297, |
| "rewards/HPSv2": 0.2669668197631836, |
| "rewards/ORM": 0.5197261422872543, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.25, |
| "step": 627 |
| }, |
| { |
| "completion_length": 57.78125, |
| "epoch": 0.6954595791805094, |
| "grad_norm": 0.7951210141181946, |
| "kl": 0.0947265625, |
| "learning_rate": 6.075e-07, |
| "loss": 0.014011627994477749, |
| "reward": 2.150346517562866, |
| "reward_std": 0.5079721957445145, |
| "rewards/GDino": 0.7525905668735504, |
| "rewards/GIT": 0.4856003522872925, |
| "rewards/HPSv2": 0.2543811798095703, |
| "rewards/ORM": 0.6577745378017426, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.4375, |
| "step": 628 |
| }, |
| { |
| "completion_length": 49.671875, |
| "epoch": 0.6965669988925803, |
| "grad_norm": 1.1048647165298462, |
| "kl": 0.102294921875, |
| "learning_rate": 6.06875e-07, |
| "loss": -0.024292804300785065, |
| "reward": 2.371622085571289, |
| "reward_std": 0.2941502034664154, |
| "rewards/GDino": 0.8161080777645111, |
| "rewards/GIT": 0.6369066536426544, |
| "rewards/HPSv2": 0.2780323028564453, |
| "rewards/ORM": 0.640575110912323, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.625, |
| "step": 629 |
| }, |
| { |
| "completion_length": 60.09375, |
| "epoch": 0.6976744186046512, |
| "grad_norm": 1.0996936559677124, |
| "kl": 0.09765625, |
| "learning_rate": 6.062499999999999e-07, |
| "loss": 6.637349724769592e-05, |
| "reward": 2.0055113434791565, |
| "reward_std": 0.1823127269744873, |
| "rewards/GDino": 0.7672508955001831, |
| "rewards/GIT": 0.34568916261196136, |
| "rewards/HPSv2": 0.2799873352050781, |
| "rewards/ORM": 0.6125839054584503, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.5, |
| "step": 630 |
| }, |
| { |
| "completion_length": 51.71875, |
| "epoch": 0.698781838316722, |
| "grad_norm": 0.8540877103805542, |
| "kl": 0.1376953125, |
| "learning_rate": 6.056249999999999e-07, |
| "loss": -0.019051258452236652, |
| "reward": 2.1809321641921997, |
| "reward_std": 0.45563268661499023, |
| "rewards/GDino": 0.75326207280159, |
| "rewards/GIT": 0.3313777446746826, |
| "rewards/HPSv2": 0.2920055389404297, |
| "rewards/ORM": 0.8042868375778198, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.5, |
| "step": 631 |
| }, |
| { |
| "completion_length": 57.015625, |
| "epoch": 0.6998892580287929, |
| "grad_norm": 0.5862644910812378, |
| "kl": 0.110595703125, |
| "learning_rate": 6.049999999999999e-07, |
| "loss": -0.0036570816300809383, |
| "reward": 1.7211298942565918, |
| "reward_std": 0.288164347410202, |
| "rewards/GDino": 0.7022069096565247, |
| "rewards/GIT": 0.29913438856601715, |
| "rewards/HPSv2": 0.2854175567626953, |
| "rewards/ORM": 0.4343710392713547, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.125, |
| "step": 632 |
| }, |
| { |
| "completion_length": 53.046875, |
| "epoch": 0.7009966777408638, |
| "grad_norm": 0.5395447015762329, |
| "kl": 0.094970703125, |
| "learning_rate": 6.043749999999999e-07, |
| "loss": 0.01028313534334302, |
| "reward": 1.823843240737915, |
| "reward_std": 0.4212976545095444, |
| "rewards/GDino": 0.6738480925559998, |
| "rewards/GIT": 0.40255512297153473, |
| "rewards/HPSv2": 0.2727527618408203, |
| "rewards/ORM": 0.47468727827072144, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.8125, |
| "step": 633 |
| }, |
| { |
| "completion_length": 63.046875, |
| "epoch": 0.7021040974529347, |
| "grad_norm": 0.5349059104919434, |
| "kl": 0.1162109375, |
| "learning_rate": 6.037499999999999e-07, |
| "loss": -0.006328664720058441, |
| "reward": 2.3039077520370483, |
| "reward_std": 0.37224359810352325, |
| "rewards/GDino": 0.8421361148357391, |
| "rewards/GIT": 0.5552612543106079, |
| "rewards/HPSv2": 0.2768268585205078, |
| "rewards/ORM": 0.6296835243701935, |
| "self_certainty_semantic": -25.5, |
| "self_certainty_token": -23.1875, |
| "step": 634 |
| }, |
| { |
| "completion_length": 64.71875, |
| "epoch": 0.7032115171650055, |
| "grad_norm": 0.6709123253822327, |
| "kl": 0.092529296875, |
| "learning_rate": 6.031249999999999e-07, |
| "loss": 0.015265015652403235, |
| "reward": 1.8476194739341736, |
| "reward_std": 0.3216380849480629, |
| "rewards/GDino": 0.6403646171092987, |
| "rewards/GIT": 0.3366498053073883, |
| "rewards/HPSv2": 0.2653312683105469, |
| "rewards/ORM": 0.6052738428115845, |
| "self_certainty_semantic": -24.875, |
| "self_certainty_token": -21.5625, |
| "step": 635 |
| }, |
| { |
| "completion_length": 59.859375, |
| "epoch": 0.7043189368770764, |
| "grad_norm": 0.6097379922866821, |
| "kl": 0.081787109375, |
| "learning_rate": 6.025000000000001e-07, |
| "loss": -0.02739726798608899, |
| "reward": 1.7737958431243896, |
| "reward_std": 0.23601312190294266, |
| "rewards/GDino": 0.665209949016571, |
| "rewards/GIT": 0.42383019626140594, |
| "rewards/HPSv2": 0.2690887451171875, |
| "rewards/ORM": 0.41566695272922516, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.375, |
| "step": 636 |
| }, |
| { |
| "completion_length": 68.96875, |
| "epoch": 0.7054263565891473, |
| "grad_norm": 1.0351394414901733, |
| "kl": 0.08056640625, |
| "learning_rate": 6.018750000000001e-07, |
| "loss": -0.01717155985534191, |
| "reward": 2.1841100454330444, |
| "reward_std": 0.2760825753211975, |
| "rewards/GDino": 0.7598958909511566, |
| "rewards/GIT": 0.5622429698705673, |
| "rewards/HPSv2": 0.28116798400878906, |
| "rewards/ORM": 0.580803394317627, |
| "self_certainty_semantic": -25.4375, |
| "self_certainty_token": -23.0625, |
| "step": 637 |
| }, |
| { |
| "completion_length": 58.4375, |
| "epoch": 0.7065337763012182, |
| "grad_norm": 0.6548821926116943, |
| "kl": 0.109375, |
| "learning_rate": 6.0125e-07, |
| "loss": -0.0038231597281992435, |
| "reward": 1.6039944291114807, |
| "reward_std": 0.45924562215805054, |
| "rewards/GDino": 0.5986713469028473, |
| "rewards/GIT": 0.23841159790754318, |
| "rewards/HPSv2": 0.25704383850097656, |
| "rewards/ORM": 0.5098675489425659, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.875, |
| "step": 638 |
| }, |
| { |
| "completion_length": 60.921875, |
| "epoch": 0.707641196013289, |
| "grad_norm": 0.8969513773918152, |
| "kl": 0.09326171875, |
| "learning_rate": 6.00625e-07, |
| "loss": -0.012395284604281187, |
| "reward": 1.9851109385490417, |
| "reward_std": 0.3337320014834404, |
| "rewards/GDino": 0.6296875178813934, |
| "rewards/GIT": 0.5730934292078018, |
| "rewards/HPSv2": 0.2747383117675781, |
| "rewards/ORM": 0.5075916647911072, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.1875, |
| "step": 639 |
| }, |
| { |
| "completion_length": 52.828125, |
| "epoch": 0.70874861572536, |
| "grad_norm": 1.3418376445770264, |
| "kl": 0.10888671875, |
| "learning_rate": 6e-07, |
| "loss": 0.001039758324623108, |
| "reward": 2.263698697090149, |
| "reward_std": 0.2723499834537506, |
| "rewards/GDino": 0.8763654232025146, |
| "rewards/GIT": 0.4212002158164978, |
| "rewards/HPSv2": 0.27863311767578125, |
| "rewards/ORM": 0.6875, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -21.5625, |
| "step": 640 |
| }, |
| { |
| "completion_length": 53.109375, |
| "epoch": 0.7098560354374308, |
| "grad_norm": 1.132075548171997, |
| "kl": 0.121337890625, |
| "learning_rate": 5.99375e-07, |
| "loss": -0.0032154805958271027, |
| "reward": 2.199480891227722, |
| "reward_std": 0.4140768200159073, |
| "rewards/GDino": 0.7215914130210876, |
| "rewards/GIT": 0.4402545690536499, |
| "rewards/HPSv2": 0.26888084411621094, |
| "rewards/ORM": 0.7687540054321289, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.5, |
| "step": 641 |
| }, |
| { |
| "completion_length": 54.59375, |
| "epoch": 0.7109634551495017, |
| "grad_norm": 0.637166440486908, |
| "kl": 0.0947265625, |
| "learning_rate": 5.9875e-07, |
| "loss": 0.017210504971444607, |
| "reward": 2.000246047973633, |
| "reward_std": 0.39111267030239105, |
| "rewards/GDino": 0.6703124940395355, |
| "rewards/GIT": 0.42524443566799164, |
| "rewards/HPSv2": 0.2597312927246094, |
| "rewards/ORM": 0.6449578106403351, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.9375, |
| "step": 642 |
| }, |
| { |
| "completion_length": 58.625, |
| "epoch": 0.7120708748615725, |
| "grad_norm": 1.12242591381073, |
| "kl": 0.1318359375, |
| "learning_rate": 5.98125e-07, |
| "loss": -0.010156976990401745, |
| "reward": 1.660601019859314, |
| "reward_std": 0.38235481083393097, |
| "rewards/GDino": 0.656350165605545, |
| "rewards/GIT": 0.22812840342521667, |
| "rewards/HPSv2": 0.2792015075683594, |
| "rewards/ORM": 0.49692094326019287, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.625, |
| "step": 643 |
| }, |
| { |
| "completion_length": 48.375, |
| "epoch": 0.7131782945736435, |
| "grad_norm": 0.8131269216537476, |
| "kl": 0.11669921875, |
| "learning_rate": 5.975e-07, |
| "loss": -0.01059816312044859, |
| "reward": 2.0672056078910828, |
| "reward_std": 0.27940696477890015, |
| "rewards/GDino": 0.6781249940395355, |
| "rewards/GIT": 0.45725762844085693, |
| "rewards/HPSv2": 0.2631568908691406, |
| "rewards/ORM": 0.6686660945415497, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -21.875, |
| "step": 644 |
| }, |
| { |
| "completion_length": 59.203125, |
| "epoch": 0.7142857142857143, |
| "grad_norm": 0.5364713072776794, |
| "kl": 0.1005859375, |
| "learning_rate": 5.96875e-07, |
| "loss": 0.008760498836636543, |
| "reward": 1.9935593605041504, |
| "reward_std": 0.2563546523451805, |
| "rewards/GDino": 0.76213538646698, |
| "rewards/GIT": 0.568911999464035, |
| "rewards/HPSv2": 0.2480487823486328, |
| "rewards/ORM": 0.4144631326198578, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -21.75, |
| "step": 645 |
| }, |
| { |
| "completion_length": 52.078125, |
| "epoch": 0.7153931339977851, |
| "grad_norm": 1.6848008632659912, |
| "kl": 0.129150390625, |
| "learning_rate": 5.962499999999999e-07, |
| "loss": 0.025898893480189145, |
| "reward": 1.969347596168518, |
| "reward_std": 0.341257244348526, |
| "rewards/GDino": 0.7203071117401123, |
| "rewards/GIT": 0.46597611904144287, |
| "rewards/HPSv2": 0.26543426513671875, |
| "rewards/ORM": 0.5176301300525665, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.0, |
| "step": 646 |
| }, |
| { |
| "completion_length": 58.109375, |
| "epoch": 0.716500553709856, |
| "grad_norm": 0.5711860060691833, |
| "kl": 0.088623046875, |
| "learning_rate": 5.956249999999999e-07, |
| "loss": 0.00951911136507988, |
| "reward": 2.052367091178894, |
| "reward_std": 0.34762145578861237, |
| "rewards/GDino": 0.8167665600776672, |
| "rewards/GIT": 0.3102717697620392, |
| "rewards/HPSv2": 0.28072166442871094, |
| "rewards/ORM": 0.6446070671081543, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.25, |
| "step": 647 |
| }, |
| { |
| "completion_length": 51.078125, |
| "epoch": 0.717607973421927, |
| "grad_norm": 0.639315664768219, |
| "kl": 0.111328125, |
| "learning_rate": 5.949999999999999e-07, |
| "loss": -0.01178273931145668, |
| "reward": 2.3945780992507935, |
| "reward_std": 0.41494397819042206, |
| "rewards/GDino": 0.8743593692779541, |
| "rewards/GIT": 0.5740520060062408, |
| "rewards/HPSv2": 0.28223419189453125, |
| "rewards/ORM": 0.663932591676712, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.4375, |
| "step": 648 |
| }, |
| { |
| "completion_length": 54.734375, |
| "epoch": 0.7187153931339978, |
| "grad_norm": 23.128345489501953, |
| "kl": 0.33935546875, |
| "learning_rate": 5.943749999999999e-07, |
| "loss": -0.02090108091942966, |
| "reward": 1.9984523057937622, |
| "reward_std": 0.38065212965011597, |
| "rewards/GDino": 0.7088541686534882, |
| "rewards/GIT": 0.2307189777493477, |
| "rewards/HPSv2": 0.28266334533691406, |
| "rewards/ORM": 0.7762157917022705, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.0625, |
| "step": 649 |
| }, |
| { |
| "completion_length": 47.84375, |
| "epoch": 0.7198228128460686, |
| "grad_norm": 0.5686013698577881, |
| "kl": 0.108642578125, |
| "learning_rate": 5.937499999999999e-07, |
| "loss": -0.022727292962372303, |
| "reward": 2.008086383342743, |
| "reward_std": 0.4491552263498306, |
| "rewards/GDino": 0.7077498733997345, |
| "rewards/GIT": 0.2243044674396515, |
| "rewards/HPSv2": 0.2886238098144531, |
| "rewards/ORM": 0.7874081134796143, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.625, |
| "step": 650 |
| }, |
| { |
| "completion_length": 60.765625, |
| "epoch": 0.7209302325581395, |
| "grad_norm": 1.299202799797058, |
| "kl": 0.111572265625, |
| "learning_rate": 5.93125e-07, |
| "loss": -0.009046670398674905, |
| "reward": 1.932914137840271, |
| "reward_std": 0.268551841378212, |
| "rewards/GDino": 0.5877524018287659, |
| "rewards/GIT": 0.428195521235466, |
| "rewards/HPSv2": 0.2728614807128906, |
| "rewards/ORM": 0.6441046893596649, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.625, |
| "step": 651 |
| }, |
| { |
| "completion_length": 46.078125, |
| "epoch": 0.7220376522702104, |
| "grad_norm": 0.6674131155014038, |
| "kl": 0.1181640625, |
| "learning_rate": 5.925e-07, |
| "loss": -0.0068924883380532265, |
| "reward": 2.38325035572052, |
| "reward_std": 0.3376559019088745, |
| "rewards/GDino": 0.7814784049987793, |
| "rewards/GIT": 0.5073752105236053, |
| "rewards/HPSv2": 0.2665424346923828, |
| "rewards/ORM": 0.827854186296463, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.6875, |
| "step": 652 |
| }, |
| { |
| "completion_length": 51.609375, |
| "epoch": 0.7231450719822813, |
| "grad_norm": 0.6701129674911499, |
| "kl": 0.111572265625, |
| "learning_rate": 5.91875e-07, |
| "loss": -0.02959446096792817, |
| "reward": 1.828068196773529, |
| "reward_std": 0.40110647678375244, |
| "rewards/GDino": 0.6408848762512207, |
| "rewards/GIT": 0.33064037561416626, |
| "rewards/HPSv2": 0.28174591064453125, |
| "rewards/ORM": 0.5747970044612885, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.375, |
| "step": 653 |
| }, |
| { |
| "completion_length": 59.296875, |
| "epoch": 0.7242524916943521, |
| "grad_norm": 0.695898711681366, |
| "kl": 0.114990234375, |
| "learning_rate": 5.912500000000001e-07, |
| "loss": -0.008497854694724083, |
| "reward": 1.9936962127685547, |
| "reward_std": 0.41329818964004517, |
| "rewards/GDino": 0.7071219682693481, |
| "rewards/GIT": 0.21919012814760208, |
| "rewards/HPSv2": 0.2739887237548828, |
| "rewards/ORM": 0.7933953106403351, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.25, |
| "step": 654 |
| }, |
| { |
| "completion_length": 55.15625, |
| "epoch": 0.7253599114064231, |
| "grad_norm": 0.6811889410018921, |
| "kl": 0.146484375, |
| "learning_rate": 5.90625e-07, |
| "loss": 0.006947606336325407, |
| "reward": 1.8089929223060608, |
| "reward_std": 0.3486994281411171, |
| "rewards/GDino": 0.7242558300495148, |
| "rewards/GIT": 0.3218558430671692, |
| "rewards/HPSv2": 0.2834606170654297, |
| "rewards/ORM": 0.4794207215309143, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.1875, |
| "step": 655 |
| }, |
| { |
| "completion_length": 58.453125, |
| "epoch": 0.7264673311184939, |
| "grad_norm": 0.5114150643348694, |
| "kl": 0.141845703125, |
| "learning_rate": 5.9e-07, |
| "loss": 0.0019571189768612385, |
| "reward": 1.85355806350708, |
| "reward_std": 0.408876970410347, |
| "rewards/GDino": 0.7061673104763031, |
| "rewards/GIT": 0.34510794281959534, |
| "rewards/HPSv2": 0.27799224853515625, |
| "rewards/ORM": 0.524290606379509, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.875, |
| "step": 656 |
| }, |
| { |
| "completion_length": 52.09375, |
| "epoch": 0.7275747508305648, |
| "grad_norm": 1.2258964776992798, |
| "kl": 0.144775390625, |
| "learning_rate": 5.89375e-07, |
| "loss": 0.0020226510241627693, |
| "reward": 2.0385664105415344, |
| "reward_std": 0.5169965326786041, |
| "rewards/GDino": 0.7201237380504608, |
| "rewards/GIT": 0.586432158946991, |
| "rewards/HPSv2": 0.2679939270019531, |
| "rewards/ORM": 0.46401654183864594, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.4375, |
| "step": 657 |
| }, |
| { |
| "completion_length": 57.71875, |
| "epoch": 0.7286821705426356, |
| "grad_norm": 1.0173580646514893, |
| "kl": 0.115234375, |
| "learning_rate": 5.8875e-07, |
| "loss": -0.014484802726656199, |
| "reward": 2.861038327217102, |
| "reward_std": 0.16512249410152435, |
| "rewards/GDino": 0.8276041150093079, |
| "rewards/GIT": 0.8040107190608978, |
| "rewards/HPSv2": 0.26067352294921875, |
| "rewards/ORM": 0.96875, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.5, |
| "step": 658 |
| }, |
| { |
| "completion_length": 61.921875, |
| "epoch": 0.7297895902547066, |
| "grad_norm": 0.6108150482177734, |
| "kl": 0.08935546875, |
| "learning_rate": 5.88125e-07, |
| "loss": -0.012970490381121635, |
| "reward": 2.240950107574463, |
| "reward_std": 0.1609238013625145, |
| "rewards/GDino": 0.8448942005634308, |
| "rewards/GIT": 0.5014997720718384, |
| "rewards/HPSv2": 0.26955604553222656, |
| "rewards/ORM": 0.625, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.625, |
| "step": 659 |
| }, |
| { |
| "completion_length": 66.59375, |
| "epoch": 0.7308970099667774, |
| "grad_norm": 0.49338072538375854, |
| "kl": 0.090576171875, |
| "learning_rate": 5.875e-07, |
| "loss": -0.00789804756641388, |
| "reward": 2.0724942684173584, |
| "reward_std": 0.3269048407673836, |
| "rewards/GDino": 0.7365051507949829, |
| "rewards/GIT": 0.5188607275485992, |
| "rewards/HPSv2": 0.2554817199707031, |
| "rewards/ORM": 0.5616468340158463, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.0625, |
| "step": 660 |
| }, |
| { |
| "completion_length": 59.296875, |
| "epoch": 0.7320044296788483, |
| "grad_norm": 0.5493965744972229, |
| "kl": 0.126708984375, |
| "learning_rate": 5.86875e-07, |
| "loss": 0.004740674514323473, |
| "reward": 1.2543038725852966, |
| "reward_std": 0.4126633256673813, |
| "rewards/GDino": 0.5057172179222107, |
| "rewards/GIT": 0.0, |
| "rewards/HPSv2": 0.2738494873046875, |
| "rewards/ORM": 0.4747372269630432, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.375, |
| "step": 661 |
| }, |
| { |
| "completion_length": 71.859375, |
| "epoch": 0.7331118493909191, |
| "grad_norm": 0.9065269231796265, |
| "kl": 0.121826171875, |
| "learning_rate": 5.8625e-07, |
| "loss": -0.019859036430716515, |
| "reward": 2.1880451440811157, |
| "reward_std": 0.43430525064468384, |
| "rewards/GDino": 0.8338541686534882, |
| "rewards/GIT": 0.5521739572286606, |
| "rewards/HPSv2": 0.2639484405517578, |
| "rewards/ORM": 0.5380684584379196, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.75, |
| "step": 662 |
| }, |
| { |
| "completion_length": 50.140625, |
| "epoch": 0.7342192691029901, |
| "grad_norm": 0.8953965306282043, |
| "kl": 0.130859375, |
| "learning_rate": 5.856249999999999e-07, |
| "loss": -0.0022711530327796936, |
| "reward": 2.8201390504837036, |
| "reward_std": 0.29404304176568985, |
| "rewards/GDino": 0.9276041686534882, |
| "rewards/GIT": 0.7852352261543274, |
| "rewards/HPSv2": 0.2791748046875, |
| "rewards/ORM": 0.8281249701976776, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.6875, |
| "step": 663 |
| }, |
| { |
| "completion_length": 43.890625, |
| "epoch": 0.7353266888150609, |
| "grad_norm": 0.6196086406707764, |
| "kl": 0.097900390625, |
| "learning_rate": 5.849999999999999e-07, |
| "loss": 0.0024156805593520403, |
| "reward": 2.224173426628113, |
| "reward_std": 0.251213937997818, |
| "rewards/GDino": 0.8485510051250458, |
| "rewards/GIT": 0.5144817233085632, |
| "rewards/HPSv2": 0.27632713317871094, |
| "rewards/ORM": 0.58481365442276, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -22.9375, |
| "step": 664 |
| }, |
| { |
| "completion_length": 59.1875, |
| "epoch": 0.7364341085271318, |
| "grad_norm": 5.437551021575928, |
| "kl": 0.58984375, |
| "learning_rate": 5.843749999999999e-07, |
| "loss": -0.012075834441930056, |
| "reward": 2.0889036655426025, |
| "reward_std": 0.4273092746734619, |
| "rewards/GDino": 0.7407647967338562, |
| "rewards/GIT": 0.31217965483665466, |
| "rewards/HPSv2": 0.2763214111328125, |
| "rewards/ORM": 0.7596378028392792, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -22.0, |
| "step": 665 |
| }, |
| { |
| "completion_length": 57.6875, |
| "epoch": 0.7375415282392026, |
| "grad_norm": 0.7798510193824768, |
| "kl": 0.10986328125, |
| "learning_rate": 5.837499999999999e-07, |
| "loss": -0.019685606472194195, |
| "reward": 2.8379937410354614, |
| "reward_std": 0.22698093205690384, |
| "rewards/GDino": 0.9088541269302368, |
| "rewards/GIT": 0.8198984265327454, |
| "rewards/HPSv2": 0.25928306579589844, |
| "rewards/ORM": 0.8499580323696136, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.5, |
| "step": 666 |
| }, |
| { |
| "completion_length": 51.140625, |
| "epoch": 0.7386489479512736, |
| "grad_norm": 0.8934307098388672, |
| "kl": 0.101806640625, |
| "learning_rate": 5.83125e-07, |
| "loss": 0.0012792428024113178, |
| "reward": 2.073898673057556, |
| "reward_std": 0.3394215703010559, |
| "rewards/GDino": 0.7093749940395355, |
| "rewards/GIT": 0.40929675102233887, |
| "rewards/HPSv2": 0.27736473083496094, |
| "rewards/ORM": 0.6778621971607208, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.0625, |
| "step": 667 |
| }, |
| { |
| "completion_length": 60.21875, |
| "epoch": 0.7397563676633444, |
| "grad_norm": 0.7199041247367859, |
| "kl": 0.101806640625, |
| "learning_rate": 5.825e-07, |
| "loss": 0.002477790229022503, |
| "reward": 2.132854461669922, |
| "reward_std": 0.34849952161312103, |
| "rewards/GDino": 0.8123269379138947, |
| "rewards/GIT": 0.4830709397792816, |
| "rewards/HPSv2": 0.27182769775390625, |
| "rewards/ORM": 0.5656289458274841, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.5625, |
| "step": 668 |
| }, |
| { |
| "completion_length": 55.09375, |
| "epoch": 0.7408637873754153, |
| "grad_norm": 0.486458420753479, |
| "kl": 0.114990234375, |
| "learning_rate": 5.81875e-07, |
| "loss": 0.0014977175742387772, |
| "reward": 2.400873064994812, |
| "reward_std": 0.32519641518592834, |
| "rewards/GDino": 0.8092281818389893, |
| "rewards/GIT": 0.5495095402002335, |
| "rewards/HPSv2": 0.2682971954345703, |
| "rewards/ORM": 0.7738381326198578, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.5625, |
| "step": 669 |
| }, |
| { |
| "completion_length": 68.6875, |
| "epoch": 0.7419712070874862, |
| "grad_norm": 0.7903674840927124, |
| "kl": 0.130859375, |
| "learning_rate": 5.8125e-07, |
| "loss": 0.017440943978726864, |
| "reward": 2.0798336267471313, |
| "reward_std": 0.30619989335536957, |
| "rewards/GDino": 0.6846449375152588, |
| "rewards/GIT": 0.3689979016780853, |
| "rewards/HPSv2": 0.27150726318359375, |
| "rewards/ORM": 0.7546834945678711, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.375, |
| "step": 670 |
| }, |
| { |
| "completion_length": 58.3125, |
| "epoch": 0.743078626799557, |
| "grad_norm": 0.7990289926528931, |
| "kl": 0.13623046875, |
| "learning_rate": 5.806249999999999e-07, |
| "loss": 0.014086478855460882, |
| "reward": 2.3666220903396606, |
| "reward_std": 0.23381435126066208, |
| "rewards/GDino": 0.7690103948116302, |
| "rewards/GIT": 0.48419545590877533, |
| "rewards/HPSv2": 0.28082847595214844, |
| "rewards/ORM": 0.8325876593589783, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.6875, |
| "step": 671 |
| }, |
| { |
| "completion_length": 57.59375, |
| "epoch": 0.7441860465116279, |
| "grad_norm": 0.7874314188957214, |
| "kl": 0.106689453125, |
| "learning_rate": 5.8e-07, |
| "loss": -0.023284037597477436, |
| "reward": 2.1920286417007446, |
| "reward_std": 0.4672909080982208, |
| "rewards/GDino": 0.7413122951984406, |
| "rewards/GIT": 0.49920889735221863, |
| "rewards/HPSv2": 0.26218223571777344, |
| "rewards/ORM": 0.6893252730369568, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.25, |
| "step": 672 |
| }, |
| { |
| "completion_length": 56.984375, |
| "epoch": 0.7452934662236987, |
| "grad_norm": 1.7885018587112427, |
| "kl": 0.134765625, |
| "learning_rate": 5.79375e-07, |
| "loss": -0.04544084332883358, |
| "reward": 1.6980464458465576, |
| "reward_std": 0.4310499578714371, |
| "rewards/GDino": 0.6950358748435974, |
| "rewards/GIT": 0.20118394494056702, |
| "rewards/HPSv2": 0.28397560119628906, |
| "rewards/ORM": 0.517850935459137, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.3125, |
| "step": 673 |
| }, |
| { |
| "completion_length": 49.796875, |
| "epoch": 0.7464008859357697, |
| "grad_norm": 0.7694756388664246, |
| "kl": 0.16015625, |
| "learning_rate": 5.7875e-07, |
| "loss": -0.009542290586978197, |
| "reward": 1.663089096546173, |
| "reward_std": 0.4869331270456314, |
| "rewards/GDino": 0.7145376801490784, |
| "rewards/GIT": 0.18890324980020523, |
| "rewards/HPSv2": 0.2716636657714844, |
| "rewards/ORM": 0.4879845529794693, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.9375, |
| "step": 674 |
| }, |
| { |
| "completion_length": 59.84375, |
| "epoch": 0.7475083056478405, |
| "grad_norm": 0.5087226629257202, |
| "kl": 0.0947265625, |
| "learning_rate": 5.78125e-07, |
| "loss": 0.012875068932771683, |
| "reward": 2.306869149208069, |
| "reward_std": 0.4066794216632843, |
| "rewards/GDino": 0.792752593755722, |
| "rewards/GIT": 0.5246446281671524, |
| "rewards/HPSv2": 0.2613048553466797, |
| "rewards/ORM": 0.7281669974327087, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.0, |
| "step": 675 |
| }, |
| { |
| "completion_length": 67.859375, |
| "epoch": 0.7486157253599114, |
| "grad_norm": 0.6212475895881653, |
| "kl": 0.135009765625, |
| "learning_rate": 5.775e-07, |
| "loss": 0.005976893939077854, |
| "reward": 2.0514276027679443, |
| "reward_std": 0.31927407532930374, |
| "rewards/GDino": 0.7584453225135803, |
| "rewards/GIT": 0.44124919176101685, |
| "rewards/HPSv2": 0.2875785827636719, |
| "rewards/ORM": 0.5641544163227081, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.9375, |
| "step": 676 |
| }, |
| { |
| "completion_length": 52.765625, |
| "epoch": 0.7497231450719822, |
| "grad_norm": 0.48944947123527527, |
| "kl": 0.1005859375, |
| "learning_rate": 5.76875e-07, |
| "loss": 0.007940568262711167, |
| "reward": 2.0665289163589478, |
| "reward_std": 0.46066904067993164, |
| "rewards/GDino": 0.764566034078598, |
| "rewards/GIT": 0.4603087604045868, |
| "rewards/HPSv2": 0.2724437713623047, |
| "rewards/ORM": 0.5692104250192642, |
| "self_certainty_semantic": -25.375, |
| "self_certainty_token": -22.5625, |
| "step": 677 |
| }, |
| { |
| "completion_length": 57.328125, |
| "epoch": 0.7508305647840532, |
| "grad_norm": 0.6796759963035583, |
| "kl": 0.104736328125, |
| "learning_rate": 5.7625e-07, |
| "loss": -0.018686085008084774, |
| "reward": 1.8018831014633179, |
| "reward_std": 0.34160397946834564, |
| "rewards/GDino": 0.7247967422008514, |
| "rewards/GIT": 0.27884362637996674, |
| "rewards/HPSv2": 0.2733306884765625, |
| "rewards/ORM": 0.5249120742082596, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -22.0625, |
| "step": 678 |
| }, |
| { |
| "completion_length": 60.375, |
| "epoch": 0.751937984496124, |
| "grad_norm": 1.11585533618927, |
| "kl": 0.14697265625, |
| "learning_rate": 5.75625e-07, |
| "loss": -0.0047287356574088335, |
| "reward": 2.0689697861671448, |
| "reward_std": 0.3384145051240921, |
| "rewards/GDino": 0.7799451053142548, |
| "rewards/GIT": 0.45028039813041687, |
| "rewards/HPSv2": 0.2711601257324219, |
| "rewards/ORM": 0.5675841867923737, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.0625, |
| "step": 679 |
| }, |
| { |
| "completion_length": 64.703125, |
| "epoch": 0.7530454042081949, |
| "grad_norm": 0.8812364935874939, |
| "kl": 0.111083984375, |
| "learning_rate": 5.749999999999999e-07, |
| "loss": -0.00965856178663671, |
| "reward": 2.2480428218841553, |
| "reward_std": 0.4152272343635559, |
| "rewards/GDino": 0.7843750417232513, |
| "rewards/GIT": 0.5545928478240967, |
| "rewards/HPSv2": 0.26071739196777344, |
| "rewards/ORM": 0.648357629776001, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.25, |
| "step": 680 |
| }, |
| { |
| "completion_length": 59.703125, |
| "epoch": 0.7541528239202658, |
| "grad_norm": 0.8357558250427246, |
| "kl": 0.1318359375, |
| "learning_rate": 5.743749999999999e-07, |
| "loss": -0.0433221198618412, |
| "reward": 2.5384799242019653, |
| "reward_std": 0.2651476636528969, |
| "rewards/GDino": 0.8453125059604645, |
| "rewards/GIT": 0.607755184173584, |
| "rewards/HPSv2": 0.25295448303222656, |
| "rewards/ORM": 0.8324578106403351, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.875, |
| "step": 681 |
| }, |
| { |
| "completion_length": 54.359375, |
| "epoch": 0.7552602436323367, |
| "grad_norm": 0.9629199504852295, |
| "kl": 0.17919921875, |
| "learning_rate": 5.737499999999999e-07, |
| "loss": -0.01920161722227931, |
| "reward": 1.8432632088661194, |
| "reward_std": 0.36826279759407043, |
| "rewards/GDino": 0.6969230771064758, |
| "rewards/GIT": 0.2167769894003868, |
| "rewards/HPSv2": 0.2827301025390625, |
| "rewards/ORM": 0.6468330323696136, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.3125, |
| "step": 682 |
| }, |
| { |
| "completion_length": 67.65625, |
| "epoch": 0.7563676633444075, |
| "grad_norm": 19.469133377075195, |
| "kl": 7.6806640625, |
| "learning_rate": 5.73125e-07, |
| "loss": 0.08776102215051651, |
| "reward": 1.6216975450515747, |
| "reward_std": 0.3734900504350662, |
| "rewards/GDino": 0.5045833885669708, |
| "rewards/GIT": 0.32963452488183975, |
| "rewards/HPSv2": 0.26216697692871094, |
| "rewards/ORM": 0.5253127440810204, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.4375, |
| "step": 683 |
| }, |
| { |
| "completion_length": 60.265625, |
| "epoch": 0.7574750830564784, |
| "grad_norm": 0.4979681372642517, |
| "kl": 0.08642578125, |
| "learning_rate": 5.725e-07, |
| "loss": -0.03171349037438631, |
| "reward": 2.428562641143799, |
| "reward_std": 0.39078929275274277, |
| "rewards/GDino": 0.7991176247596741, |
| "rewards/GIT": 0.6331989169120789, |
| "rewards/HPSv2": 0.2715587615966797, |
| "rewards/ORM": 0.724687248468399, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.6875, |
| "step": 684 |
| }, |
| { |
| "completion_length": 57.125, |
| "epoch": 0.7585825027685493, |
| "grad_norm": 0.5635210275650024, |
| "kl": 0.1220703125, |
| "learning_rate": 5.71875e-07, |
| "loss": 0.007911482825875282, |
| "reward": 1.8947407603263855, |
| "reward_std": 0.5111456513404846, |
| "rewards/GDino": 0.7101004719734192, |
| "rewards/GIT": 0.42421260476112366, |
| "rewards/HPSv2": 0.27680397033691406, |
| "rewards/ORM": 0.483623668551445, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.625, |
| "step": 685 |
| }, |
| { |
| "completion_length": 58.21875, |
| "epoch": 0.7596899224806202, |
| "grad_norm": 0.7649754285812378, |
| "kl": 0.102294921875, |
| "learning_rate": 5.7125e-07, |
| "loss": -0.0001670820638537407, |
| "reward": 1.9881569147109985, |
| "reward_std": 0.22512035071849823, |
| "rewards/GDino": 0.6786458194255829, |
| "rewards/GIT": 0.4509306624531746, |
| "rewards/HPSv2": 0.26122093200683594, |
| "rewards/ORM": 0.5973595678806305, |
| "self_certainty_semantic": -25.125, |
| "self_certainty_token": -22.0625, |
| "step": 686 |
| }, |
| { |
| "completion_length": 58.609375, |
| "epoch": 0.760797342192691, |
| "grad_norm": 0.7712872624397278, |
| "kl": 0.150390625, |
| "learning_rate": 5.70625e-07, |
| "loss": -0.01166512817144394, |
| "reward": 2.13641893863678, |
| "reward_std": 0.3073776960372925, |
| "rewards/GDino": 0.7664711773395538, |
| "rewards/GIT": 0.5652618110179901, |
| "rewards/HPSv2": 0.2640190124511719, |
| "rewards/ORM": 0.5406669676303864, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.3125, |
| "step": 687 |
| }, |
| { |
| "completion_length": 77.484375, |
| "epoch": 0.7619047619047619, |
| "grad_norm": 0.7627017498016357, |
| "kl": 0.1142578125, |
| "learning_rate": 5.699999999999999e-07, |
| "loss": -0.0010195523500442505, |
| "reward": 2.0781058073043823, |
| "reward_std": 0.3603600710630417, |
| "rewards/GDino": 0.7742711901664734, |
| "rewards/GIT": 0.4729154258966446, |
| "rewards/HPSv2": 0.2698936462402344, |
| "rewards/ORM": 0.5610254108905792, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -21.8125, |
| "step": 688 |
| }, |
| { |
| "completion_length": 64.765625, |
| "epoch": 0.7630121816168328, |
| "grad_norm": 0.6321325898170471, |
| "kl": 0.102783203125, |
| "learning_rate": 5.69375e-07, |
| "loss": 0.019116760464385152, |
| "reward": 1.90728360414505, |
| "reward_std": 0.3873142600059509, |
| "rewards/GDino": 0.7718750238418579, |
| "rewards/GIT": 0.467089980840683, |
| "rewards/HPSv2": 0.2593402862548828, |
| "rewards/ORM": 0.4089783579111099, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.625, |
| "step": 689 |
| }, |
| { |
| "completion_length": 70.53125, |
| "epoch": 0.7641196013289037, |
| "grad_norm": 0.7649714350700378, |
| "kl": 0.1396484375, |
| "learning_rate": 5.6875e-07, |
| "loss": -0.010147146880626678, |
| "reward": 2.009207248687744, |
| "reward_std": 0.5445281267166138, |
| "rewards/GDino": 0.7276042103767395, |
| "rewards/GIT": 0.4141872376203537, |
| "rewards/HPSv2": 0.27762603759765625, |
| "rewards/ORM": 0.5897897481918335, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -21.75, |
| "step": 690 |
| }, |
| { |
| "completion_length": 66.0, |
| "epoch": 0.7652270210409745, |
| "grad_norm": 0.7068275213241577, |
| "kl": 0.1337890625, |
| "learning_rate": 5.68125e-07, |
| "loss": 0.03244199859909713, |
| "reward": 1.9956420063972473, |
| "reward_std": 0.25323009490966797, |
| "rewards/GDino": 0.6093957275152206, |
| "rewards/GIT": 0.40899063646793365, |
| "rewards/HPSv2": 0.2871570587158203, |
| "rewards/ORM": 0.6900985240936279, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.1875, |
| "step": 691 |
| }, |
| { |
| "completion_length": 63.03125, |
| "epoch": 0.7663344407530454, |
| "grad_norm": 823.07958984375, |
| "kl": 33.553466796875, |
| "learning_rate": 5.675e-07, |
| "loss": 0.35094713792204857, |
| "reward": 1.9486876726150513, |
| "reward_std": 0.3541366904973984, |
| "rewards/GDino": 0.7794270813465118, |
| "rewards/GIT": 0.551772803068161, |
| "rewards/HPSv2": 0.2665576934814453, |
| "rewards/ORM": 0.35093018412590027, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.6875, |
| "step": 692 |
| }, |
| { |
| "completion_length": 51.171875, |
| "epoch": 0.7674418604651163, |
| "grad_norm": 0.6784785389900208, |
| "kl": 0.150390625, |
| "learning_rate": 5.66875e-07, |
| "loss": 0.0222349907271564, |
| "reward": 2.4065383076667786, |
| "reward_std": 0.42152607440948486, |
| "rewards/GDino": 0.8412966132164001, |
| "rewards/GIT": 0.4581999033689499, |
| "rewards/HPSv2": 0.2733802795410156, |
| "rewards/ORM": 0.8336616158485413, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.625, |
| "step": 693 |
| }, |
| { |
| "completion_length": 61.53125, |
| "epoch": 0.7685492801771872, |
| "grad_norm": 0.5948299169540405, |
| "kl": 0.1083984375, |
| "learning_rate": 5.6625e-07, |
| "loss": -0.01655739266425371, |
| "reward": 1.9052525162696838, |
| "reward_std": 0.4417829215526581, |
| "rewards/GDino": 0.6542519927024841, |
| "rewards/GIT": 0.33645208925008774, |
| "rewards/HPSv2": 0.26097679138183594, |
| "rewards/ORM": 0.653571605682373, |
| "self_certainty_semantic": -25.25, |
| "self_certainty_token": -22.25, |
| "step": 694 |
| }, |
| { |
| "completion_length": 69.953125, |
| "epoch": 0.769656699889258, |
| "grad_norm": 0.5376792550086975, |
| "kl": 0.094482421875, |
| "learning_rate": 5.65625e-07, |
| "loss": -0.021405136212706566, |
| "reward": 2.114523470401764, |
| "reward_std": 0.3266918957233429, |
| "rewards/GDino": 0.7836482524871826, |
| "rewards/GIT": 0.5422925800085068, |
| "rewards/HPSv2": 0.2541236877441406, |
| "rewards/ORM": 0.5344589352607727, |
| "self_certainty_semantic": -25.0, |
| "self_certainty_token": -22.3125, |
| "step": 695 |
| }, |
| { |
| "completion_length": 65.390625, |
| "epoch": 0.770764119601329, |
| "grad_norm": 1.5156573057174683, |
| "kl": 0.154296875, |
| "learning_rate": 5.649999999999999e-07, |
| "loss": -0.012026362586766481, |
| "reward": 1.7989888191223145, |
| "reward_std": 0.46254193782806396, |
| "rewards/GDino": 0.7475877106189728, |
| "rewards/GIT": 0.21283076703548431, |
| "rewards/HPSv2": 0.2808036804199219, |
| "rewards/ORM": 0.5577665567398071, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.3125, |
| "step": 696 |
| }, |
| { |
| "completion_length": 62.71875, |
| "epoch": 0.7718715393133998, |
| "grad_norm": 0.9091273546218872, |
| "kl": 0.1533203125, |
| "learning_rate": 5.643749999999999e-07, |
| "loss": 0.0035165061708539724, |
| "reward": 2.0271888971328735, |
| "reward_std": 0.43045464158058167, |
| "rewards/GDino": 0.6859623789787292, |
| "rewards/GIT": 0.4345651715993881, |
| "rewards/HPSv2": 0.25859642028808594, |
| "rewards/ORM": 0.6480649709701538, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -21.875, |
| "step": 697 |
| }, |
| { |
| "completion_length": 54.671875, |
| "epoch": 0.7729789590254706, |
| "grad_norm": 0.8914833664894104, |
| "kl": 0.12939453125, |
| "learning_rate": 5.637499999999999e-07, |
| "loss": -0.013794566038995981, |
| "reward": 1.7404972314834595, |
| "reward_std": 0.355168879032135, |
| "rewards/GDino": 0.6277631521224976, |
| "rewards/GIT": 0.31849637627601624, |
| "rewards/HPSv2": 0.2751331329345703, |
| "rewards/ORM": 0.5191046893596649, |
| "self_certainty_semantic": -25.0625, |
| "self_certainty_token": -22.0625, |
| "step": 698 |
| }, |
| { |
| "completion_length": 58.5625, |
| "epoch": 0.7740863787375415, |
| "grad_norm": 1.0078788995742798, |
| "kl": 0.13720703125, |
| "learning_rate": 5.63125e-07, |
| "loss": 0.022074274718761444, |
| "reward": 2.429232120513916, |
| "reward_std": 0.40295615792274475, |
| "rewards/GDino": 0.7623399794101715, |
| "rewards/GIT": 0.5622873604297638, |
| "rewards/HPSv2": 0.27647972106933594, |
| "rewards/ORM": 0.8281249701976776, |
| "self_certainty_semantic": -25.3125, |
| "self_certainty_token": -22.6875, |
| "step": 699 |
| }, |
| { |
| "completion_length": 54.078125, |
| "epoch": 0.7751937984496124, |
| "grad_norm": 0.600312352180481, |
| "kl": 0.1435546875, |
| "learning_rate": 5.625e-07, |
| "loss": -0.010840186383575201, |
| "reward": 1.9293212890625, |
| "reward_std": 0.4079190194606781, |
| "rewards/GDino": 0.6971202790737152, |
| "rewards/GIT": 0.41506989300251007, |
| "rewards/HPSv2": 0.26502037048339844, |
| "rewards/ORM": 0.552110880613327, |
| "self_certainty_semantic": -25.1875, |
| "self_certainty_token": -22.3125, |
| "step": 700 |
| } |
| ], |
| "logging_steps": 1.0, |
| "max_steps": 1600, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 100, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
