| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.8213552361396304, |
| "eval_steps": 500, |
| "global_step": 1000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.9327721463143825, |
| "epoch": 0.02053388090349076, |
| "grad_norm": 0.1806640625, |
| "learning_rate": 0.00012972972972972974, |
| "loss": 0.9855536651611329, |
| "mean_token_accuracy": 0.7679368060827255, |
| "num_tokens": 201640.0, |
| "step": 25 |
| }, |
| { |
| "entropy": 0.8724879887700081, |
| "epoch": 0.04106776180698152, |
| "grad_norm": 0.1474609375, |
| "learning_rate": 0.00019994905570360817, |
| "loss": 0.8708956909179687, |
| "mean_token_accuracy": 0.7903838059306145, |
| "num_tokens": 404792.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.8544691229611635, |
| "epoch": 0.061601642710472276, |
| "grad_norm": 0.15234375, |
| "learning_rate": 0.00019951602516690988, |
| "loss": 0.8482267761230469, |
| "mean_token_accuracy": 0.7932471469044685, |
| "num_tokens": 606810.0, |
| "step": 75 |
| }, |
| { |
| "entropy": 0.8944924513995648, |
| "epoch": 0.08213552361396304, |
| "grad_norm": 0.1640625, |
| "learning_rate": 0.0001986430349043317, |
| "loss": 0.8818518829345703, |
| "mean_token_accuracy": 0.7834139719605446, |
| "num_tokens": 807549.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.9099837489426136, |
| "epoch": 0.1026694045174538, |
| "grad_norm": 0.14453125, |
| "learning_rate": 0.0001973339444003735, |
| "loss": 0.8984813690185547, |
| "mean_token_accuracy": 0.7812782523036003, |
| "num_tokens": 1008223.0, |
| "step": 125 |
| }, |
| { |
| "entropy": 0.8356300877034664, |
| "epoch": 0.12320328542094455, |
| "grad_norm": 0.15234375, |
| "learning_rate": 0.0001955945411359792, |
| "loss": 0.8289723205566406, |
| "mean_token_accuracy": 0.7977875503897667, |
| "num_tokens": 1209811.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.8626699298620224, |
| "epoch": 0.1437371663244353, |
| "grad_norm": 0.150390625, |
| "learning_rate": 0.00019343251500211975, |
| "loss": 0.8522069549560547, |
| "mean_token_accuracy": 0.789976358115673, |
| "num_tokens": 1411353.0, |
| "step": 175 |
| }, |
| { |
| "entropy": 0.8640103512257338, |
| "epoch": 0.16427104722792607, |
| "grad_norm": 0.1318359375, |
| "learning_rate": 0.00019085742430283322, |
| "loss": 0.8584527587890625, |
| "mean_token_accuracy": 0.7891600033640862, |
| "num_tokens": 1614555.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.8809252215921879, |
| "epoch": 0.18480492813141683, |
| "grad_norm": 0.1640625, |
| "learning_rate": 0.0001878806534980221, |
| "loss": 0.87360107421875, |
| "mean_token_accuracy": 0.7867547670006751, |
| "num_tokens": 1816672.0, |
| "step": 225 |
| }, |
| { |
| "entropy": 0.8541410507261753, |
| "epoch": 0.2053388090349076, |
| "grad_norm": 0.11962890625, |
| "learning_rate": 0.0001845153628728274, |
| "loss": 0.8500338745117187, |
| "mean_token_accuracy": 0.7926329699158668, |
| "num_tokens": 2018500.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.8946534705162048, |
| "epoch": 0.22587268993839835, |
| "grad_norm": 0.1689453125, |
| "learning_rate": 0.00018077643035609004, |
| "loss": 0.8828852844238281, |
| "mean_token_accuracy": 0.7846844819188118, |
| "num_tokens": 2220866.0, |
| "step": 275 |
| }, |
| { |
| "entropy": 0.8587514898180961, |
| "epoch": 0.2464065708418891, |
| "grad_norm": 0.171875, |
| "learning_rate": 0.0001766803857451204, |
| "loss": 0.8391613006591797, |
| "mean_token_accuracy": 0.7931238517165184, |
| "num_tokens": 2421013.0, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.876440706178546, |
| "epoch": 0.2669404517453799, |
| "grad_norm": 0.1796875, |
| "learning_rate": 0.00017224533762756776, |
| "loss": 0.8661175537109375, |
| "mean_token_accuracy": 0.7875098848342895, |
| "num_tokens": 2623081.0, |
| "step": 325 |
| }, |
| { |
| "entropy": 0.8844135162234307, |
| "epoch": 0.2874743326488706, |
| "grad_norm": 0.15234375, |
| "learning_rate": 0.00016749089332346713, |
| "loss": 0.8735830688476562, |
| "mean_token_accuracy": 0.7862337788939476, |
| "num_tokens": 2824382.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.7987543179094792, |
| "epoch": 0.3080082135523614, |
| "grad_norm": 0.154296875, |
| "learning_rate": 0.00016243807220139988, |
| "loss": 0.79259033203125, |
| "mean_token_accuracy": 0.804975683093071, |
| "num_tokens": 3026700.0, |
| "step": 375 |
| }, |
| { |
| "entropy": 0.8895229130983353, |
| "epoch": 0.32854209445585214, |
| "grad_norm": 0.171875, |
| "learning_rate": 0.0001571092127519967, |
| "loss": 0.8765971374511718, |
| "mean_token_accuracy": 0.7862004506587982, |
| "num_tokens": 3228741.0, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.8609522663801908, |
| "epoch": 0.3490759753593429, |
| "grad_norm": 0.1484375, |
| "learning_rate": 0.00015152787382960968, |
| "loss": 0.8444467163085938, |
| "mean_token_accuracy": 0.793184621334076, |
| "num_tokens": 3428336.0, |
| "step": 425 |
| }, |
| { |
| "entropy": 0.8780751578509808, |
| "epoch": 0.36960985626283366, |
| "grad_norm": 0.1318359375, |
| "learning_rate": 0.0001457187304987645, |
| "loss": 0.8594348907470704, |
| "mean_token_accuracy": 0.7891285745799541, |
| "num_tokens": 3626721.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.815336870700121, |
| "epoch": 0.39014373716632444, |
| "grad_norm": 0.1357421875, |
| "learning_rate": 0.0001397074649458544, |
| "loss": 0.8068878936767578, |
| "mean_token_accuracy": 0.8028223878145218, |
| "num_tokens": 3829062.0, |
| "step": 475 |
| }, |
| { |
| "entropy": 0.8639824625849724, |
| "epoch": 0.4106776180698152, |
| "grad_norm": 0.169921875, |
| "learning_rate": 0.00013352065293835397, |
| "loss": 0.8533393096923828, |
| "mean_token_accuracy": 0.7899997106194496, |
| "num_tokens": 4030678.0, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.8904591681063175, |
| "epoch": 0.43121149897330596, |
| "grad_norm": 0.162109375, |
| "learning_rate": 0.00012718564633351772, |
| "loss": 0.8847498321533203, |
| "mean_token_accuracy": 0.7833931747078896, |
| "num_tokens": 4233119.0, |
| "step": 525 |
| }, |
| { |
| "entropy": 0.8268530152738094, |
| "epoch": 0.4517453798767967, |
| "grad_norm": 0.1689453125, |
| "learning_rate": 0.00012073045215598952, |
| "loss": 0.8184004211425782, |
| "mean_token_accuracy": 0.7992048686742783, |
| "num_tokens": 4434734.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.7941572253406047, |
| "epoch": 0.4722792607802875, |
| "grad_norm": 0.173828125, |
| "learning_rate": 0.00011418360877892164, |
| "loss": 0.7798463439941407, |
| "mean_token_accuracy": 0.8061937466263771, |
| "num_tokens": 4636778.0, |
| "step": 575 |
| }, |
| { |
| "entropy": 0.8667953862249851, |
| "epoch": 0.4928131416837782, |
| "grad_norm": 0.169921875, |
| "learning_rate": 0.00010757405975600534, |
| "loss": 0.8635111999511719, |
| "mean_token_accuracy": 0.7889971971511841, |
| "num_tokens": 4839819.0, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.8342785247415304, |
| "epoch": 0.5133470225872689, |
| "grad_norm": 0.1416015625, |
| "learning_rate": 0.00010093102586220055, |
| "loss": 0.8236666870117187, |
| "mean_token_accuracy": 0.796975427120924, |
| "num_tokens": 5041753.0, |
| "step": 625 |
| }, |
| { |
| "entropy": 0.8055999881774187, |
| "epoch": 0.5338809034907598, |
| "grad_norm": 0.158203125, |
| "learning_rate": 9.42838759088723e-05, |
| "loss": 0.7989521789550781, |
| "mean_token_accuracy": 0.805096383690834, |
| "num_tokens": 5245040.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.8180603013932705, |
| "epoch": 0.5544147843942505, |
| "grad_norm": 0.150390625, |
| "learning_rate": 8.76619969044592e-05, |
| "loss": 0.802100601196289, |
| "mean_token_accuracy": 0.8017301180958748, |
| "num_tokens": 5446302.0, |
| "step": 675 |
| }, |
| { |
| "entropy": 0.8475391804426908, |
| "epoch": 0.5749486652977412, |
| "grad_norm": 0.1865234375, |
| "learning_rate": 8.109466413469238e-05, |
| "loss": 0.8385409545898438, |
| "mean_token_accuracy": 0.7944941192865371, |
| "num_tokens": 5647189.0, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.8185767906159163, |
| "epoch": 0.5954825462012321, |
| "grad_norm": 0.201171875, |
| "learning_rate": 7.461091173674022e-05, |
| "loss": 0.80953369140625, |
| "mean_token_accuracy": 0.7994823551177979, |
| "num_tokens": 5849070.0, |
| "step": 725 |
| }, |
| { |
| "entropy": 0.8287316697835923, |
| "epoch": 0.6160164271047228, |
| "grad_norm": 0.193359375, |
| "learning_rate": 6.823940433946921e-05, |
| "loss": 0.8172183227539063, |
| "mean_token_accuracy": 0.8001478153467179, |
| "num_tokens": 6051514.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.8862237848341465, |
| "epoch": 0.6365503080082136, |
| "grad_norm": 0.1630859375, |
| "learning_rate": 6.200831033729863e-05, |
| "loss": 0.8690660095214844, |
| "mean_token_accuracy": 0.7873855569958687, |
| "num_tokens": 6251375.0, |
| "step": 775 |
| }, |
| { |
| "entropy": 0.7810994558781386, |
| "epoch": 0.6570841889117043, |
| "grad_norm": 0.1826171875, |
| "learning_rate": 5.5945177357904934e-05, |
| "loss": 0.7692386627197265, |
| "mean_token_accuracy": 0.8092155620455742, |
| "num_tokens": 6453040.0, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.8132312067598104, |
| "epoch": 0.6776180698151951, |
| "grad_norm": 0.1572265625, |
| "learning_rate": 5.007681047433139e-05, |
| "loss": 0.8110353088378907, |
| "mean_token_accuracy": 0.8015294006466865, |
| "num_tokens": 6656782.0, |
| "step": 825 |
| }, |
| { |
| "entropy": 0.8274586813151836, |
| "epoch": 0.6981519507186859, |
| "grad_norm": 0.1630859375, |
| "learning_rate": 4.442915369992802e-05, |
| "loss": 0.815575942993164, |
| "mean_token_accuracy": 0.7989534091949463, |
| "num_tokens": 6858454.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.8976877976208926, |
| "epoch": 0.7186858316221766, |
| "grad_norm": 0.162109375, |
| "learning_rate": 3.902717529003005e-05, |
| "loss": 0.8837800598144532, |
| "mean_token_accuracy": 0.7825979651510715, |
| "num_tokens": 7058361.0, |
| "step": 875 |
| }, |
| { |
| "entropy": 0.8282124460488558, |
| "epoch": 0.7392197125256673, |
| "grad_norm": 0.1640625, |
| "learning_rate": 3.389475735745761e-05, |
| "loss": 0.8223667907714843, |
| "mean_token_accuracy": 0.79724593937397, |
| "num_tokens": 7261833.0, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.8559306030720473, |
| "epoch": 0.7597535934291582, |
| "grad_norm": 0.1630859375, |
| "learning_rate": 2.9054590289843396e-05, |
| "loss": 0.8454450988769531, |
| "mean_token_accuracy": 0.7930123561620712, |
| "num_tokens": 7463682.0, |
| "step": 925 |
| }, |
| { |
| "entropy": 0.8133819183707237, |
| "epoch": 0.7802874743326489, |
| "grad_norm": 0.1884765625, |
| "learning_rate": 2.452807243557116e-05, |
| "loss": 0.8075272369384766, |
| "mean_token_accuracy": 0.8024582426249981, |
| "num_tokens": 7666583.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.8377064400911332, |
| "epoch": 0.8008213552361396, |
| "grad_norm": 0.1748046875, |
| "learning_rate": 2.0335215501810777e-05, |
| "loss": 0.825996322631836, |
| "mean_token_accuracy": 0.7960993593931198, |
| "num_tokens": 7867378.0, |
| "step": 975 |
| }, |
| { |
| "entropy": 0.8585446905344725, |
| "epoch": 0.8213552361396304, |
| "grad_norm": 0.169921875, |
| "learning_rate": 1.6494556082886036e-05, |
| "loss": 0.8439271545410156, |
| "mean_token_accuracy": 0.7945261257886886, |
| "num_tokens": 8068940.0, |
| "step": 1000 |
| } |
| ], |
| "logging_steps": 25, |
| "max_steps": 1218, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 6.807092753482138e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
