output_0216_2e6_all / checkpoint-1000 /trainer_state.json
Prummn's picture
Add files using upload-large-folder tool
c9c6229 verified
{
"best_metric": 0.43363777,
"best_model_checkpoint": "/data/liuzihang/haobin/pangkaiyu/output/output_step_audio2_mini-encoder+align+llm-whole0130_signal_new1_dpdc-lora-1gpu-bs16_4_gckF_2e6_all/v2-20260215-150801/checkpoint-1000",
"epoch": 0.2109704641350211,
"eval_steps": 200,
"global_step": 1000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0002109704641350211,
"grad_norm": 0.76171875,
"learning_rate": 5.263157894736842e-09,
"loss": 0.2671268880367279,
"step": 1,
"token_acc": 0.9268792044457443
},
{
"epoch": 0.0004219409282700422,
"grad_norm": 0.6640625,
"learning_rate": 1.0526315789473684e-08,
"loss": 0.25691983103752136,
"step": 2,
"token_acc": 0.9315326633165829
},
{
"epoch": 0.0006329113924050633,
"grad_norm": 0.58984375,
"learning_rate": 1.5789473684210525e-08,
"loss": 0.20553478598594666,
"step": 3,
"token_acc": 0.9399263247378861
},
{
"epoch": 0.0008438818565400844,
"grad_norm": 0.7578125,
"learning_rate": 2.1052631578947368e-08,
"loss": 0.2874465584754944,
"step": 4,
"token_acc": 0.9190901238122661
},
{
"epoch": 0.0010548523206751054,
"grad_norm": 0.99609375,
"learning_rate": 2.6315789473684208e-08,
"loss": 0.23955097794532776,
"step": 5,
"token_acc": 0.9297171186934966
},
{
"epoch": 0.0012658227848101266,
"grad_norm": 0.6484375,
"learning_rate": 3.157894736842105e-08,
"loss": 0.22314852476119995,
"step": 6,
"token_acc": 0.9365693865396069
},
{
"epoch": 0.0014767932489451476,
"grad_norm": 0.74609375,
"learning_rate": 3.684210526315789e-08,
"loss": 0.23751771450042725,
"step": 7,
"token_acc": 0.9336440910337264
},
{
"epoch": 0.0016877637130801688,
"grad_norm": 0.86328125,
"learning_rate": 4.2105263157894737e-08,
"loss": 0.22823776304721832,
"step": 8,
"token_acc": 0.9342301943198804
},
{
"epoch": 0.0018987341772151898,
"grad_norm": 0.6015625,
"learning_rate": 4.736842105263158e-08,
"loss": 0.22278322279453278,
"step": 9,
"token_acc": 0.9367531331973186
},
{
"epoch": 0.002109704641350211,
"grad_norm": 0.72265625,
"learning_rate": 5.2631578947368416e-08,
"loss": 0.25690096616744995,
"step": 10,
"token_acc": 0.9291217257318952
},
{
"epoch": 0.002320675105485232,
"grad_norm": 1.046875,
"learning_rate": 5.789473684210526e-08,
"loss": 0.3066456913948059,
"step": 11,
"token_acc": 0.9201006605850897
},
{
"epoch": 0.002531645569620253,
"grad_norm": 0.7890625,
"learning_rate": 6.31578947368421e-08,
"loss": 0.2525354027748108,
"step": 12,
"token_acc": 0.9276785714285715
},
{
"epoch": 0.0027426160337552744,
"grad_norm": 0.76953125,
"learning_rate": 6.842105263157895e-08,
"loss": 0.30059731006622314,
"step": 13,
"token_acc": 0.9238062986793092
},
{
"epoch": 0.002953586497890295,
"grad_norm": 0.9921875,
"learning_rate": 7.368421052631577e-08,
"loss": 0.250629723072052,
"step": 14,
"token_acc": 0.9291187739463601
},
{
"epoch": 0.0031645569620253164,
"grad_norm": 0.6796875,
"learning_rate": 7.894736842105262e-08,
"loss": 0.25745123624801636,
"step": 15,
"token_acc": 0.9317230273752013
},
{
"epoch": 0.0033755274261603376,
"grad_norm": 0.85546875,
"learning_rate": 8.421052631578947e-08,
"loss": 0.3376100957393646,
"step": 16,
"token_acc": 0.9077240566037735
},
{
"epoch": 0.003586497890295359,
"grad_norm": 0.703125,
"learning_rate": 8.947368421052631e-08,
"loss": 0.24760955572128296,
"step": 17,
"token_acc": 0.9222253760999148
},
{
"epoch": 0.0037974683544303796,
"grad_norm": 0.703125,
"learning_rate": 9.473684210526316e-08,
"loss": 0.2046602964401245,
"step": 18,
"token_acc": 0.9403993855606759
},
{
"epoch": 0.004008438818565401,
"grad_norm": 0.83203125,
"learning_rate": 1e-07,
"loss": 0.26379868388175964,
"step": 19,
"token_acc": 0.9294478527607362
},
{
"epoch": 0.004219409282700422,
"grad_norm": 0.72265625,
"learning_rate": 1.0526315789473683e-07,
"loss": 0.2657994031906128,
"step": 20,
"token_acc": 0.9254714157437893
},
{
"epoch": 0.004430379746835443,
"grad_norm": 0.6640625,
"learning_rate": 1.1052631578947368e-07,
"loss": 0.28728997707366943,
"step": 21,
"token_acc": 0.9262518968133535
},
{
"epoch": 0.004641350210970464,
"grad_norm": 0.77734375,
"learning_rate": 1.1578947368421052e-07,
"loss": 0.27618926763534546,
"step": 22,
"token_acc": 0.9308072487644151
},
{
"epoch": 0.004852320675105486,
"grad_norm": 0.7109375,
"learning_rate": 1.2105263157894737e-07,
"loss": 0.2314767688512802,
"step": 23,
"token_acc": 0.936124911284599
},
{
"epoch": 0.005063291139240506,
"grad_norm": 0.73828125,
"learning_rate": 1.263157894736842e-07,
"loss": 0.24274200201034546,
"step": 24,
"token_acc": 0.9345039018952063
},
{
"epoch": 0.005274261603375527,
"grad_norm": 0.64453125,
"learning_rate": 1.3157894736842104e-07,
"loss": 0.2632070481777191,
"step": 25,
"token_acc": 0.9223796033994335
},
{
"epoch": 0.005485232067510549,
"grad_norm": 0.7421875,
"learning_rate": 1.368421052631579e-07,
"loss": 0.2736364006996155,
"step": 26,
"token_acc": 0.9269776876267748
},
{
"epoch": 0.00569620253164557,
"grad_norm": 0.96875,
"learning_rate": 1.4210526315789474e-07,
"loss": 0.29377132654190063,
"step": 27,
"token_acc": 0.9181763285024155
},
{
"epoch": 0.00590717299578059,
"grad_norm": 0.7734375,
"learning_rate": 1.4736842105263155e-07,
"loss": 0.25689125061035156,
"step": 28,
"token_acc": 0.9299856527977044
},
{
"epoch": 0.006118143459915612,
"grad_norm": 0.9609375,
"learning_rate": 1.526315789473684e-07,
"loss": 0.24775874614715576,
"step": 29,
"token_acc": 0.9330016583747927
},
{
"epoch": 0.006329113924050633,
"grad_norm": 0.703125,
"learning_rate": 1.5789473684210525e-07,
"loss": 0.25338542461395264,
"step": 30,
"token_acc": 0.925512104283054
},
{
"epoch": 0.006540084388185654,
"grad_norm": 0.6328125,
"learning_rate": 1.631578947368421e-07,
"loss": 0.25087809562683105,
"step": 31,
"token_acc": 0.9313120472229676
},
{
"epoch": 0.006751054852320675,
"grad_norm": 0.67578125,
"learning_rate": 1.6842105263157895e-07,
"loss": 0.2502059042453766,
"step": 32,
"token_acc": 0.9279077218840115
},
{
"epoch": 0.006962025316455696,
"grad_norm": 0.578125,
"learning_rate": 1.7368421052631578e-07,
"loss": 0.18295930325984955,
"step": 33,
"token_acc": 0.9424480628860191
},
{
"epoch": 0.007172995780590718,
"grad_norm": 0.7265625,
"learning_rate": 1.7894736842105262e-07,
"loss": 0.2690507471561432,
"step": 34,
"token_acc": 0.9241155819605725
},
{
"epoch": 0.007383966244725738,
"grad_norm": 0.90234375,
"learning_rate": 1.8421052631578946e-07,
"loss": 0.2535433769226074,
"step": 35,
"token_acc": 0.9346446700507615
},
{
"epoch": 0.007594936708860759,
"grad_norm": 0.73046875,
"learning_rate": 1.8947368421052632e-07,
"loss": 0.26006314158439636,
"step": 36,
"token_acc": 0.9291425420457678
},
{
"epoch": 0.007805907172995781,
"grad_norm": 0.86328125,
"learning_rate": 1.9473684210526315e-07,
"loss": 0.2664929926395416,
"step": 37,
"token_acc": 0.9286128845037724
},
{
"epoch": 0.008016877637130802,
"grad_norm": 0.65234375,
"learning_rate": 2e-07,
"loss": 0.2170935869216919,
"step": 38,
"token_acc": 0.9359218028780885
},
{
"epoch": 0.008227848101265823,
"grad_norm": 0.75390625,
"learning_rate": 2.0526315789473683e-07,
"loss": 0.31706634163856506,
"step": 39,
"token_acc": 0.9133278822567457
},
{
"epoch": 0.008438818565400843,
"grad_norm": 0.6953125,
"learning_rate": 2.1052631578947366e-07,
"loss": 0.23433184623718262,
"step": 40,
"token_acc": 0.9313227829202747
},
{
"epoch": 0.008649789029535865,
"grad_norm": 0.65234375,
"learning_rate": 2.1578947368421053e-07,
"loss": 0.19157642126083374,
"step": 41,
"token_acc": 0.9416859122401847
},
{
"epoch": 0.008860759493670886,
"grad_norm": 0.75390625,
"learning_rate": 2.2105263157894736e-07,
"loss": 0.26239246129989624,
"step": 42,
"token_acc": 0.9240352476450927
},
{
"epoch": 0.009071729957805906,
"grad_norm": 0.97265625,
"learning_rate": 2.263157894736842e-07,
"loss": 0.27333155274391174,
"step": 43,
"token_acc": 0.9244935543278084
},
{
"epoch": 0.009282700421940928,
"grad_norm": 0.671875,
"learning_rate": 2.3157894736842104e-07,
"loss": 0.22739389538764954,
"step": 44,
"token_acc": 0.9345622119815669
},
{
"epoch": 0.00949367088607595,
"grad_norm": 1.359375,
"learning_rate": 2.3684210526315787e-07,
"loss": 0.2970912754535675,
"step": 45,
"token_acc": 0.918966119455117
},
{
"epoch": 0.009704641350210971,
"grad_norm": 0.98046875,
"learning_rate": 2.4210526315789473e-07,
"loss": 0.24367359280586243,
"step": 46,
"token_acc": 0.9307627357162961
},
{
"epoch": 0.009915611814345991,
"grad_norm": 0.69140625,
"learning_rate": 2.4736842105263157e-07,
"loss": 0.24166589975357056,
"step": 47,
"token_acc": 0.925096985974336
},
{
"epoch": 0.010126582278481013,
"grad_norm": 0.8203125,
"learning_rate": 2.526315789473684e-07,
"loss": 0.2498053014278412,
"step": 48,
"token_acc": 0.9309855154785572
},
{
"epoch": 0.010337552742616034,
"grad_norm": 0.81640625,
"learning_rate": 2.578947368421053e-07,
"loss": 0.27882808446884155,
"step": 49,
"token_acc": 0.9232902033271719
},
{
"epoch": 0.010548523206751054,
"grad_norm": 0.7890625,
"learning_rate": 2.631578947368421e-07,
"loss": 0.2516263723373413,
"step": 50,
"token_acc": 0.9283572142619126
},
{
"epoch": 0.010759493670886076,
"grad_norm": 0.65625,
"learning_rate": 2.684210526315789e-07,
"loss": 0.22138270735740662,
"step": 51,
"token_acc": 0.9392942583732058
},
{
"epoch": 0.010970464135021098,
"grad_norm": 1.015625,
"learning_rate": 2.736842105263158e-07,
"loss": 0.25101763010025024,
"step": 52,
"token_acc": 0.9323397913561848
},
{
"epoch": 0.011181434599156118,
"grad_norm": 0.77734375,
"learning_rate": 2.789473684210526e-07,
"loss": 0.2590043842792511,
"step": 53,
"token_acc": 0.9248989023685731
},
{
"epoch": 0.01139240506329114,
"grad_norm": 1.03125,
"learning_rate": 2.842105263157895e-07,
"loss": 0.22629833221435547,
"step": 54,
"token_acc": 0.9388133498145859
},
{
"epoch": 0.011603375527426161,
"grad_norm": 1.2421875,
"learning_rate": 2.894736842105263e-07,
"loss": 0.26315873861312866,
"step": 55,
"token_acc": 0.9215632686526374
},
{
"epoch": 0.01181434599156118,
"grad_norm": 1.390625,
"learning_rate": 2.947368421052631e-07,
"loss": 0.3269142806529999,
"step": 56,
"token_acc": 0.9163541967118546
},
{
"epoch": 0.012025316455696202,
"grad_norm": 0.8125,
"learning_rate": 3e-07,
"loss": 0.2740277945995331,
"step": 57,
"token_acc": 0.9223241590214067
},
{
"epoch": 0.012236286919831224,
"grad_norm": 0.7109375,
"learning_rate": 3.052631578947368e-07,
"loss": 0.2807028889656067,
"step": 58,
"token_acc": 0.9231628946633138
},
{
"epoch": 0.012447257383966244,
"grad_norm": 0.83984375,
"learning_rate": 3.105263157894737e-07,
"loss": 0.3021116256713867,
"step": 59,
"token_acc": 0.9209953343701399
},
{
"epoch": 0.012658227848101266,
"grad_norm": 0.73828125,
"learning_rate": 3.157894736842105e-07,
"loss": 0.2364785373210907,
"step": 60,
"token_acc": 0.9329750237116662
},
{
"epoch": 0.012869198312236287,
"grad_norm": 0.7109375,
"learning_rate": 3.2105263157894733e-07,
"loss": 0.2884541153907776,
"step": 61,
"token_acc": 0.9274515831540117
},
{
"epoch": 0.013080168776371307,
"grad_norm": 0.76953125,
"learning_rate": 3.263157894736842e-07,
"loss": 0.25490057468414307,
"step": 62,
"token_acc": 0.9336579427875837
},
{
"epoch": 0.013291139240506329,
"grad_norm": 0.859375,
"learning_rate": 3.31578947368421e-07,
"loss": 0.27591922879219055,
"step": 63,
"token_acc": 0.9201367308887508
},
{
"epoch": 0.01350210970464135,
"grad_norm": 0.83203125,
"learning_rate": 3.368421052631579e-07,
"loss": 0.2646903693675995,
"step": 64,
"token_acc": 0.9261189454322502
},
{
"epoch": 0.013713080168776372,
"grad_norm": 0.7265625,
"learning_rate": 3.4210526315789473e-07,
"loss": 0.2481774091720581,
"step": 65,
"token_acc": 0.9300921512551636
},
{
"epoch": 0.013924050632911392,
"grad_norm": 0.671875,
"learning_rate": 3.4736842105263157e-07,
"loss": 0.2667776644229889,
"step": 66,
"token_acc": 0.9202678027997565
},
{
"epoch": 0.014135021097046414,
"grad_norm": 0.77734375,
"learning_rate": 3.526315789473684e-07,
"loss": 0.2720962464809418,
"step": 67,
"token_acc": 0.9237356168049238
},
{
"epoch": 0.014345991561181435,
"grad_norm": 0.6171875,
"learning_rate": 3.5789473684210524e-07,
"loss": 0.25555452704429626,
"step": 68,
"token_acc": 0.9306469298245614
},
{
"epoch": 0.014556962025316455,
"grad_norm": 0.640625,
"learning_rate": 3.6315789473684213e-07,
"loss": 0.22453869879245758,
"step": 69,
"token_acc": 0.9388444990780578
},
{
"epoch": 0.014767932489451477,
"grad_norm": 0.75,
"learning_rate": 3.684210526315789e-07,
"loss": 0.28728553652763367,
"step": 70,
"token_acc": 0.9182989690721649
},
{
"epoch": 0.014978902953586498,
"grad_norm": 0.95703125,
"learning_rate": 3.7368421052631575e-07,
"loss": 0.2622889578342438,
"step": 71,
"token_acc": 0.9239098624524437
},
{
"epoch": 0.015189873417721518,
"grad_norm": 0.78125,
"learning_rate": 3.7894736842105264e-07,
"loss": 0.2780531346797943,
"step": 72,
"token_acc": 0.9266362252663622
},
{
"epoch": 0.01540084388185654,
"grad_norm": 0.703125,
"learning_rate": 3.842105263157894e-07,
"loss": 0.2625043988227844,
"step": 73,
"token_acc": 0.9331468531468532
},
{
"epoch": 0.015611814345991562,
"grad_norm": 0.703125,
"learning_rate": 3.894736842105263e-07,
"loss": 0.25795984268188477,
"step": 74,
"token_acc": 0.9242160278745645
},
{
"epoch": 0.015822784810126583,
"grad_norm": 0.71484375,
"learning_rate": 3.9473684210526315e-07,
"loss": 0.2481173276901245,
"step": 75,
"token_acc": 0.9310846176214016
},
{
"epoch": 0.016033755274261603,
"grad_norm": 0.640625,
"learning_rate": 4e-07,
"loss": 0.23631326854228973,
"step": 76,
"token_acc": 0.9369342184671092
},
{
"epoch": 0.016244725738396623,
"grad_norm": 0.69140625,
"learning_rate": 4.052631578947368e-07,
"loss": 0.24659401178359985,
"step": 77,
"token_acc": 0.930279458369346
},
{
"epoch": 0.016455696202531647,
"grad_norm": 0.77734375,
"learning_rate": 4.1052631578947365e-07,
"loss": 0.28330034017562866,
"step": 78,
"token_acc": 0.9203966005665722
},
{
"epoch": 0.016666666666666666,
"grad_norm": 0.96484375,
"learning_rate": 4.1578947368421054e-07,
"loss": 0.2582593560218811,
"step": 79,
"token_acc": 0.9296250768285187
},
{
"epoch": 0.016877637130801686,
"grad_norm": 0.734375,
"learning_rate": 4.2105263157894733e-07,
"loss": 0.2518593370914459,
"step": 80,
"token_acc": 0.9288208434058555
},
{
"epoch": 0.01708860759493671,
"grad_norm": 0.859375,
"learning_rate": 4.2631578947368416e-07,
"loss": 0.30441492795944214,
"step": 81,
"token_acc": 0.9242614707730987
},
{
"epoch": 0.01729957805907173,
"grad_norm": 0.7578125,
"learning_rate": 4.3157894736842105e-07,
"loss": 0.30916911363601685,
"step": 82,
"token_acc": 0.9124253625248792
},
{
"epoch": 0.01751054852320675,
"grad_norm": 0.609375,
"learning_rate": 4.368421052631579e-07,
"loss": 0.28638702630996704,
"step": 83,
"token_acc": 0.9245689655172413
},
{
"epoch": 0.017721518987341773,
"grad_norm": 0.80859375,
"learning_rate": 4.421052631578947e-07,
"loss": 0.2646373510360718,
"step": 84,
"token_acc": 0.9279176201372997
},
{
"epoch": 0.017932489451476793,
"grad_norm": 0.97265625,
"learning_rate": 4.4736842105263156e-07,
"loss": 0.27530571818351746,
"step": 85,
"token_acc": 0.9217687074829932
},
{
"epoch": 0.018143459915611813,
"grad_norm": 0.73828125,
"learning_rate": 4.526315789473684e-07,
"loss": 0.30989915132522583,
"step": 86,
"token_acc": 0.9133137062479555
},
{
"epoch": 0.018354430379746836,
"grad_norm": 0.91015625,
"learning_rate": 4.5789473684210523e-07,
"loss": 0.2850973308086395,
"step": 87,
"token_acc": 0.9211218229623137
},
{
"epoch": 0.018565400843881856,
"grad_norm": 0.72265625,
"learning_rate": 4.6315789473684207e-07,
"loss": 0.2523067593574524,
"step": 88,
"token_acc": 0.9337899543378996
},
{
"epoch": 0.018776371308016876,
"grad_norm": 0.734375,
"learning_rate": 4.6842105263157896e-07,
"loss": 0.2510542869567871,
"step": 89,
"token_acc": 0.9345845983991168
},
{
"epoch": 0.0189873417721519,
"grad_norm": 0.65625,
"learning_rate": 4.7368421052631574e-07,
"loss": 0.27831587195396423,
"step": 90,
"token_acc": 0.9237072619384007
},
{
"epoch": 0.01919831223628692,
"grad_norm": 0.67578125,
"learning_rate": 4.789473684210526e-07,
"loss": 0.2806670665740967,
"step": 91,
"token_acc": 0.9245553643144004
},
{
"epoch": 0.019409282700421943,
"grad_norm": 0.8046875,
"learning_rate": 4.842105263157895e-07,
"loss": 0.2630135416984558,
"step": 92,
"token_acc": 0.9274447949526814
},
{
"epoch": 0.019620253164556962,
"grad_norm": 0.71875,
"learning_rate": 4.894736842105263e-07,
"loss": 0.2662945091724396,
"step": 93,
"token_acc": 0.9303838646714379
},
{
"epoch": 0.019831223628691982,
"grad_norm": 0.703125,
"learning_rate": 4.947368421052631e-07,
"loss": 0.2430751472711563,
"step": 94,
"token_acc": 0.9333737129012719
},
{
"epoch": 0.020042194092827006,
"grad_norm": 0.80078125,
"learning_rate": 5e-07,
"loss": 0.2676389515399933,
"step": 95,
"token_acc": 0.9231901118304885
},
{
"epoch": 0.020253164556962026,
"grad_norm": 0.765625,
"learning_rate": 5.052631578947368e-07,
"loss": 0.24123789370059967,
"step": 96,
"token_acc": 0.9292089873807325
},
{
"epoch": 0.020464135021097046,
"grad_norm": 0.80078125,
"learning_rate": 5.105263157894736e-07,
"loss": 0.24845993518829346,
"step": 97,
"token_acc": 0.9311657879320445
},
{
"epoch": 0.02067510548523207,
"grad_norm": 0.6953125,
"learning_rate": 5.157894736842106e-07,
"loss": 0.2502240836620331,
"step": 98,
"token_acc": 0.9316136772645471
},
{
"epoch": 0.02088607594936709,
"grad_norm": 0.64453125,
"learning_rate": 5.210526315789473e-07,
"loss": 0.2715577483177185,
"step": 99,
"token_acc": 0.9284542172628816
},
{
"epoch": 0.02109704641350211,
"grad_norm": 0.640625,
"learning_rate": 5.263157894736842e-07,
"loss": 0.25573915243148804,
"step": 100,
"token_acc": 0.9303937007874016
},
{
"epoch": 0.021308016877637132,
"grad_norm": 0.5859375,
"learning_rate": 5.31578947368421e-07,
"loss": 0.23634707927703857,
"step": 101,
"token_acc": 0.9278404618210443
},
{
"epoch": 0.021518987341772152,
"grad_norm": 0.79296875,
"learning_rate": 5.368421052631578e-07,
"loss": 0.24541448056697845,
"step": 102,
"token_acc": 0.9269794721407625
},
{
"epoch": 0.021729957805907172,
"grad_norm": 0.8046875,
"learning_rate": 5.421052631578948e-07,
"loss": 0.29984721541404724,
"step": 103,
"token_acc": 0.9175257731958762
},
{
"epoch": 0.021940928270042195,
"grad_norm": 0.703125,
"learning_rate": 5.473684210526316e-07,
"loss": 0.23752300441265106,
"step": 104,
"token_acc": 0.9307146753955264
},
{
"epoch": 0.022151898734177215,
"grad_norm": 0.7578125,
"learning_rate": 5.526315789473684e-07,
"loss": 0.26209786534309387,
"step": 105,
"token_acc": 0.9331191588785047
},
{
"epoch": 0.022362869198312235,
"grad_norm": 0.6015625,
"learning_rate": 5.578947368421052e-07,
"loss": 0.26135504245758057,
"step": 106,
"token_acc": 0.9239098624524437
},
{
"epoch": 0.02257383966244726,
"grad_norm": 0.76171875,
"learning_rate": 5.63157894736842e-07,
"loss": 0.2286645919084549,
"step": 107,
"token_acc": 0.9346016646848989
},
{
"epoch": 0.02278481012658228,
"grad_norm": 0.94921875,
"learning_rate": 5.68421052631579e-07,
"loss": 0.2228844314813614,
"step": 108,
"token_acc": 0.9335610058987892
},
{
"epoch": 0.0229957805907173,
"grad_norm": 0.7265625,
"learning_rate": 5.736842105263158e-07,
"loss": 0.23979443311691284,
"step": 109,
"token_acc": 0.9332460732984293
},
{
"epoch": 0.023206751054852322,
"grad_norm": 0.7734375,
"learning_rate": 5.789473684210526e-07,
"loss": 0.2925586402416229,
"step": 110,
"token_acc": 0.9247163973874184
},
{
"epoch": 0.02341772151898734,
"grad_norm": 0.89453125,
"learning_rate": 5.842105263157895e-07,
"loss": 0.26546645164489746,
"step": 111,
"token_acc": 0.9265167364016736
},
{
"epoch": 0.02362869198312236,
"grad_norm": 0.71875,
"learning_rate": 5.894736842105262e-07,
"loss": 0.23430243134498596,
"step": 112,
"token_acc": 0.9325294286534597
},
{
"epoch": 0.023839662447257385,
"grad_norm": 0.6875,
"learning_rate": 5.947368421052631e-07,
"loss": 0.2555156946182251,
"step": 113,
"token_acc": 0.9277472527472528
},
{
"epoch": 0.024050632911392405,
"grad_norm": 0.625,
"learning_rate": 6e-07,
"loss": 0.27733132243156433,
"step": 114,
"token_acc": 0.9281454979129398
},
{
"epoch": 0.024261603375527425,
"grad_norm": 0.6328125,
"learning_rate": 6.052631578947368e-07,
"loss": 0.2438090741634369,
"step": 115,
"token_acc": 0.9304691916336914
},
{
"epoch": 0.024472573839662448,
"grad_norm": 0.70703125,
"learning_rate": 6.105263157894736e-07,
"loss": 0.2810766100883484,
"step": 116,
"token_acc": 0.9260921603830042
},
{
"epoch": 0.024683544303797468,
"grad_norm": 0.6171875,
"learning_rate": 6.157894736842105e-07,
"loss": 0.21388083696365356,
"step": 117,
"token_acc": 0.9422394320748628
},
{
"epoch": 0.024894514767932488,
"grad_norm": 0.8828125,
"learning_rate": 6.210526315789474e-07,
"loss": 0.304085373878479,
"step": 118,
"token_acc": 0.9148753224419605
},
{
"epoch": 0.02510548523206751,
"grad_norm": 0.8125,
"learning_rate": 6.263157894736842e-07,
"loss": 0.24785375595092773,
"step": 119,
"token_acc": 0.9345686160972785
},
{
"epoch": 0.02531645569620253,
"grad_norm": 0.6015625,
"learning_rate": 6.31578947368421e-07,
"loss": 0.23662757873535156,
"step": 120,
"token_acc": 0.9339464882943144
},
{
"epoch": 0.02552742616033755,
"grad_norm": 0.7890625,
"learning_rate": 6.368421052631578e-07,
"loss": 0.30741703510284424,
"step": 121,
"token_acc": 0.9190517490604221
},
{
"epoch": 0.025738396624472575,
"grad_norm": 0.58984375,
"learning_rate": 6.421052631578947e-07,
"loss": 0.2324860692024231,
"step": 122,
"token_acc": 0.9350512753089666
},
{
"epoch": 0.025949367088607594,
"grad_norm": 0.71875,
"learning_rate": 6.473684210526316e-07,
"loss": 0.22093364596366882,
"step": 123,
"token_acc": 0.9382284382284383
},
{
"epoch": 0.026160337552742614,
"grad_norm": 0.58203125,
"learning_rate": 6.526315789473684e-07,
"loss": 0.26719486713409424,
"step": 124,
"token_acc": 0.9285504263451926
},
{
"epoch": 0.026371308016877638,
"grad_norm": 0.69921875,
"learning_rate": 6.578947368421053e-07,
"loss": 0.26403629779815674,
"step": 125,
"token_acc": 0.9270715096481271
},
{
"epoch": 0.026582278481012658,
"grad_norm": 1.046875,
"learning_rate": 6.63157894736842e-07,
"loss": 0.24764738976955414,
"step": 126,
"token_acc": 0.9328358208955224
},
{
"epoch": 0.02679324894514768,
"grad_norm": 0.80078125,
"learning_rate": 6.684210526315788e-07,
"loss": 0.2644532322883606,
"step": 127,
"token_acc": 0.9267714201008005
},
{
"epoch": 0.0270042194092827,
"grad_norm": 0.80859375,
"learning_rate": 6.736842105263158e-07,
"loss": 0.2985777258872986,
"step": 128,
"token_acc": 0.9214629997164729
},
{
"epoch": 0.02721518987341772,
"grad_norm": 0.62109375,
"learning_rate": 6.789473684210526e-07,
"loss": 0.23605869710445404,
"step": 129,
"token_acc": 0.9335576114381834
},
{
"epoch": 0.027426160337552744,
"grad_norm": 0.72265625,
"learning_rate": 6.842105263157895e-07,
"loss": 0.254613995552063,
"step": 130,
"token_acc": 0.9289265867212635
},
{
"epoch": 0.027637130801687764,
"grad_norm": 0.91015625,
"learning_rate": 6.894736842105263e-07,
"loss": 0.32649004459381104,
"step": 131,
"token_acc": 0.9107457428068115
},
{
"epoch": 0.027848101265822784,
"grad_norm": 0.66796875,
"learning_rate": 6.947368421052631e-07,
"loss": 0.2345716655254364,
"step": 132,
"token_acc": 0.9308671922377199
},
{
"epoch": 0.028059071729957807,
"grad_norm": 0.796875,
"learning_rate": 7e-07,
"loss": 0.2658767104148865,
"step": 133,
"token_acc": 0.9226377390807879
},
{
"epoch": 0.028270042194092827,
"grad_norm": 0.70703125,
"learning_rate": 7.052631578947368e-07,
"loss": 0.2791082561016083,
"step": 134,
"token_acc": 0.9298836497244336
},
{
"epoch": 0.028481012658227847,
"grad_norm": 0.75390625,
"learning_rate": 7.105263157894736e-07,
"loss": 0.28342780470848083,
"step": 135,
"token_acc": 0.9173340961098398
},
{
"epoch": 0.02869198312236287,
"grad_norm": 0.53125,
"learning_rate": 7.157894736842105e-07,
"loss": 0.19060048460960388,
"step": 136,
"token_acc": 0.9426000620539869
},
{
"epoch": 0.02890295358649789,
"grad_norm": 0.76171875,
"learning_rate": 7.210526315789473e-07,
"loss": 0.24044275283813477,
"step": 137,
"token_acc": 0.9315665883931566
},
{
"epoch": 0.02911392405063291,
"grad_norm": 0.69921875,
"learning_rate": 7.263157894736843e-07,
"loss": 0.22750994563102722,
"step": 138,
"token_acc": 0.9367798193709125
},
{
"epoch": 0.029324894514767934,
"grad_norm": 0.65234375,
"learning_rate": 7.315789473684211e-07,
"loss": 0.2409280240535736,
"step": 139,
"token_acc": 0.9357879234167894
},
{
"epoch": 0.029535864978902954,
"grad_norm": 0.74609375,
"learning_rate": 7.368421052631578e-07,
"loss": 0.24030432105064392,
"step": 140,
"token_acc": 0.928141912206855
},
{
"epoch": 0.029746835443037974,
"grad_norm": 0.703125,
"learning_rate": 7.421052631578947e-07,
"loss": 0.255402147769928,
"step": 141,
"token_acc": 0.9247853124074622
},
{
"epoch": 0.029957805907172997,
"grad_norm": 0.74609375,
"learning_rate": 7.473684210526315e-07,
"loss": 0.21290147304534912,
"step": 142,
"token_acc": 0.9362466327446872
},
{
"epoch": 0.030168776371308017,
"grad_norm": 0.88671875,
"learning_rate": 7.526315789473684e-07,
"loss": 0.2536450922489166,
"step": 143,
"token_acc": 0.9314112291350531
},
{
"epoch": 0.030379746835443037,
"grad_norm": 0.6015625,
"learning_rate": 7.578947368421053e-07,
"loss": 0.20192307233810425,
"step": 144,
"token_acc": 0.9473519272955186
},
{
"epoch": 0.03059071729957806,
"grad_norm": 0.7734375,
"learning_rate": 7.631578947368421e-07,
"loss": 0.27158498764038086,
"step": 145,
"token_acc": 0.9229805886036319
},
{
"epoch": 0.03080168776371308,
"grad_norm": 0.93359375,
"learning_rate": 7.684210526315788e-07,
"loss": 0.2483355551958084,
"step": 146,
"token_acc": 0.9249113760876571
},
{
"epoch": 0.0310126582278481,
"grad_norm": 0.546875,
"learning_rate": 7.736842105263157e-07,
"loss": 0.2523292303085327,
"step": 147,
"token_acc": 0.9309432853364679
},
{
"epoch": 0.031223628691983123,
"grad_norm": 0.9375,
"learning_rate": 7.789473684210526e-07,
"loss": 0.28833281993865967,
"step": 148,
"token_acc": 0.9170937594211637
},
{
"epoch": 0.03143459915611815,
"grad_norm": 0.7421875,
"learning_rate": 7.842105263157895e-07,
"loss": 0.250460147857666,
"step": 149,
"token_acc": 0.9285714285714286
},
{
"epoch": 0.03164556962025317,
"grad_norm": 0.9453125,
"learning_rate": 7.894736842105263e-07,
"loss": 0.2840534746646881,
"step": 150,
"token_acc": 0.9216349108789182
},
{
"epoch": 0.03185654008438819,
"grad_norm": 0.74609375,
"learning_rate": 7.947368421052631e-07,
"loss": 0.2967279851436615,
"step": 151,
"token_acc": 0.9164603960396039
},
{
"epoch": 0.032067510548523206,
"grad_norm": 0.83203125,
"learning_rate": 8e-07,
"loss": 0.25097131729125977,
"step": 152,
"token_acc": 0.929299572509043
},
{
"epoch": 0.032278481012658226,
"grad_norm": 0.6953125,
"learning_rate": 8.052631578947368e-07,
"loss": 0.23201878368854523,
"step": 153,
"token_acc": 0.9383025367992484
},
{
"epoch": 0.032489451476793246,
"grad_norm": 0.65625,
"learning_rate": 8.105263157894736e-07,
"loss": 0.2245524525642395,
"step": 154,
"token_acc": 0.9406554472984943
},
{
"epoch": 0.03270042194092827,
"grad_norm": 0.75,
"learning_rate": 8.157894736842105e-07,
"loss": 0.22958879172801971,
"step": 155,
"token_acc": 0.9288455860643637
},
{
"epoch": 0.03291139240506329,
"grad_norm": 0.70703125,
"learning_rate": 8.210526315789473e-07,
"loss": 0.2539287805557251,
"step": 156,
"token_acc": 0.9255386565272496
},
{
"epoch": 0.03312236286919831,
"grad_norm": 0.6875,
"learning_rate": 8.263157894736841e-07,
"loss": 0.245978444814682,
"step": 157,
"token_acc": 0.9341576506955178
},
{
"epoch": 0.03333333333333333,
"grad_norm": 0.66796875,
"learning_rate": 8.315789473684211e-07,
"loss": 0.26281115412712097,
"step": 158,
"token_acc": 0.927784222737819
},
{
"epoch": 0.03354430379746835,
"grad_norm": 1.0,
"learning_rate": 8.368421052631579e-07,
"loss": 0.229181706905365,
"step": 159,
"token_acc": 0.9329073482428115
},
{
"epoch": 0.03375527426160337,
"grad_norm": 0.73046875,
"learning_rate": 8.421052631578947e-07,
"loss": 0.2598130702972412,
"step": 160,
"token_acc": 0.9302388707926167
},
{
"epoch": 0.0339662447257384,
"grad_norm": 0.75,
"learning_rate": 8.473684210526315e-07,
"loss": 0.25113117694854736,
"step": 161,
"token_acc": 0.9281177829099307
},
{
"epoch": 0.03417721518987342,
"grad_norm": 0.6015625,
"learning_rate": 8.526315789473683e-07,
"loss": 0.20510195195674896,
"step": 162,
"token_acc": 0.936978417266187
},
{
"epoch": 0.03438818565400844,
"grad_norm": 0.75,
"learning_rate": 8.578947368421053e-07,
"loss": 0.25259485840797424,
"step": 163,
"token_acc": 0.9289311695579183
},
{
"epoch": 0.03459915611814346,
"grad_norm": 0.6484375,
"learning_rate": 8.631578947368421e-07,
"loss": 0.2823118567466736,
"step": 164,
"token_acc": 0.9225122349102773
},
{
"epoch": 0.03481012658227848,
"grad_norm": 0.84375,
"learning_rate": 8.684210526315789e-07,
"loss": 0.2576562762260437,
"step": 165,
"token_acc": 0.931045050566963
},
{
"epoch": 0.0350210970464135,
"grad_norm": 0.6796875,
"learning_rate": 8.736842105263158e-07,
"loss": 0.2349683940410614,
"step": 166,
"token_acc": 0.9357770372614359
},
{
"epoch": 0.035232067510548526,
"grad_norm": 0.89453125,
"learning_rate": 8.789473684210525e-07,
"loss": 0.19951120018959045,
"step": 167,
"token_acc": 0.9381918819188192
},
{
"epoch": 0.035443037974683546,
"grad_norm": 0.58984375,
"learning_rate": 8.842105263157895e-07,
"loss": 0.24796079099178314,
"step": 168,
"token_acc": 0.9311111111111111
},
{
"epoch": 0.035654008438818566,
"grad_norm": 0.73828125,
"learning_rate": 8.894736842105263e-07,
"loss": 0.2532733082771301,
"step": 169,
"token_acc": 0.9297912713472486
},
{
"epoch": 0.035864978902953586,
"grad_norm": 0.98046875,
"learning_rate": 8.947368421052631e-07,
"loss": 0.25062763690948486,
"step": 170,
"token_acc": 0.9289099526066351
},
{
"epoch": 0.036075949367088606,
"grad_norm": 0.6484375,
"learning_rate": 9e-07,
"loss": 0.2228512465953827,
"step": 171,
"token_acc": 0.9302030456852792
},
{
"epoch": 0.036286919831223625,
"grad_norm": 0.55859375,
"learning_rate": 9.052631578947368e-07,
"loss": 0.20684444904327393,
"step": 172,
"token_acc": 0.9388583019414662
},
{
"epoch": 0.03649789029535865,
"grad_norm": 0.68359375,
"learning_rate": 9.105263157894737e-07,
"loss": 0.2119678407907486,
"step": 173,
"token_acc": 0.938135593220339
},
{
"epoch": 0.03670886075949367,
"grad_norm": 0.828125,
"learning_rate": 9.157894736842105e-07,
"loss": 0.25168293714523315,
"step": 174,
"token_acc": 0.9275818639798489
},
{
"epoch": 0.03691983122362869,
"grad_norm": 2.046875,
"learning_rate": 9.210526315789473e-07,
"loss": 0.2518053650856018,
"step": 175,
"token_acc": 0.9275232105420784
},
{
"epoch": 0.03713080168776371,
"grad_norm": 0.6171875,
"learning_rate": 9.263157894736841e-07,
"loss": 0.25257354974746704,
"step": 176,
"token_acc": 0.9272777932571747
},
{
"epoch": 0.03734177215189873,
"grad_norm": 1.0390625,
"learning_rate": 9.31578947368421e-07,
"loss": 0.24908028542995453,
"step": 177,
"token_acc": 0.9316982303632412
},
{
"epoch": 0.03755274261603375,
"grad_norm": 0.796875,
"learning_rate": 9.368421052631579e-07,
"loss": 0.2594815194606781,
"step": 178,
"token_acc": 0.9222846441947565
},
{
"epoch": 0.03776371308016878,
"grad_norm": 1.734375,
"learning_rate": 9.421052631578948e-07,
"loss": 0.301219642162323,
"step": 179,
"token_acc": 0.9261637239165329
},
{
"epoch": 0.0379746835443038,
"grad_norm": 0.6171875,
"learning_rate": 9.473684210526315e-07,
"loss": 0.2224687933921814,
"step": 180,
"token_acc": 0.9298196166854565
},
{
"epoch": 0.03818565400843882,
"grad_norm": 0.76953125,
"learning_rate": 9.526315789473683e-07,
"loss": 0.2755109667778015,
"step": 181,
"token_acc": 0.9245460237946149
},
{
"epoch": 0.03839662447257384,
"grad_norm": 0.6796875,
"learning_rate": 9.578947368421053e-07,
"loss": 0.24350810050964355,
"step": 182,
"token_acc": 0.9347500748278958
},
{
"epoch": 0.03860759493670886,
"grad_norm": 0.70703125,
"learning_rate": 9.63157894736842e-07,
"loss": 0.26835542917251587,
"step": 183,
"token_acc": 0.9246247205365697
},
{
"epoch": 0.038818565400843885,
"grad_norm": 0.703125,
"learning_rate": 9.68421052631579e-07,
"loss": 0.25252771377563477,
"step": 184,
"token_acc": 0.9330130016958734
},
{
"epoch": 0.039029535864978905,
"grad_norm": 1.21875,
"learning_rate": 9.736842105263158e-07,
"loss": 0.27294090390205383,
"step": 185,
"token_acc": 0.9220917822838848
},
{
"epoch": 0.039240506329113925,
"grad_norm": 0.9140625,
"learning_rate": 9.789473684210526e-07,
"loss": 0.257973313331604,
"step": 186,
"token_acc": 0.9298401420959147
},
{
"epoch": 0.039451476793248945,
"grad_norm": 0.71875,
"learning_rate": 9.842105263157894e-07,
"loss": 0.20286661386489868,
"step": 187,
"token_acc": 0.9379893517068587
},
{
"epoch": 0.039662447257383965,
"grad_norm": 1.0859375,
"learning_rate": 9.894736842105263e-07,
"loss": 0.30547526478767395,
"step": 188,
"token_acc": 0.9224360815857512
},
{
"epoch": 0.039873417721518985,
"grad_norm": 0.6796875,
"learning_rate": 9.947368421052631e-07,
"loss": 0.2581551671028137,
"step": 189,
"token_acc": 0.9263346257083209
},
{
"epoch": 0.04008438818565401,
"grad_norm": 0.99609375,
"learning_rate": 1e-06,
"loss": 0.267391562461853,
"step": 190,
"token_acc": 0.9272947591638897
},
{
"epoch": 0.04029535864978903,
"grad_norm": 0.7265625,
"learning_rate": 9.99999971410384e-07,
"loss": 0.22360718250274658,
"step": 191,
"token_acc": 0.9367160775370581
},
{
"epoch": 0.04050632911392405,
"grad_norm": 0.71875,
"learning_rate": 9.999998856415392e-07,
"loss": 0.2589290738105774,
"step": 192,
"token_acc": 0.9256432004523607
},
{
"epoch": 0.04071729957805907,
"grad_norm": 0.65234375,
"learning_rate": 9.999997426934757e-07,
"loss": 0.2469128668308258,
"step": 193,
"token_acc": 0.9290098745663197
},
{
"epoch": 0.04092827004219409,
"grad_norm": 1.203125,
"learning_rate": 9.999995425662095e-07,
"loss": 0.2602100968360901,
"step": 194,
"token_acc": 0.9284731774415406
},
{
"epoch": 0.04113924050632911,
"grad_norm": 0.61328125,
"learning_rate": 9.999992852597638e-07,
"loss": 0.2579442858695984,
"step": 195,
"token_acc": 0.9280116110304789
},
{
"epoch": 0.04135021097046414,
"grad_norm": 0.75,
"learning_rate": 9.999989707741678e-07,
"loss": 0.266757071018219,
"step": 196,
"token_acc": 0.923786841321822
},
{
"epoch": 0.04156118143459916,
"grad_norm": 0.59765625,
"learning_rate": 9.999985991094577e-07,
"loss": 0.22912907600402832,
"step": 197,
"token_acc": 0.9336933693369337
},
{
"epoch": 0.04177215189873418,
"grad_norm": 0.73046875,
"learning_rate": 9.999981702656756e-07,
"loss": 0.23296543955802917,
"step": 198,
"token_acc": 0.931110498759989
},
{
"epoch": 0.0419831223628692,
"grad_norm": 0.75390625,
"learning_rate": 9.999976842428708e-07,
"loss": 0.27944594621658325,
"step": 199,
"token_acc": 0.9281785829828535
},
{
"epoch": 0.04219409282700422,
"grad_norm": 0.6328125,
"learning_rate": 9.99997141041099e-07,
"loss": 0.2449186146259308,
"step": 200,
"token_acc": 0.9309120699071546
},
{
"epoch": 0.04219409282700422,
"eval_loss": 0.43372446298599243,
"eval_runtime": 245.8313,
"eval_samples_per_second": 137.106,
"eval_steps_per_second": 2.144,
"eval_token_acc": 0.8990801399982051,
"step": 200
},
{
"epoch": 0.04240506329113924,
"grad_norm": 0.765625,
"learning_rate": 9.99996540660422e-07,
"loss": 0.2534567713737488,
"step": 201,
"token_acc": 0.9282550930026572
},
{
"epoch": 0.042616033755274264,
"grad_norm": 1.0,
"learning_rate": 9.999958831009087e-07,
"loss": 0.2861325144767761,
"step": 202,
"token_acc": 0.9250295159386068
},
{
"epoch": 0.042827004219409284,
"grad_norm": 0.671875,
"learning_rate": 9.999951683626345e-07,
"loss": 0.24760206043720245,
"step": 203,
"token_acc": 0.9320360151031077
},
{
"epoch": 0.043037974683544304,
"grad_norm": 0.62109375,
"learning_rate": 9.999943964456805e-07,
"loss": 0.2488883137702942,
"step": 204,
"token_acc": 0.9300783604581073
},
{
"epoch": 0.043248945147679324,
"grad_norm": 0.80078125,
"learning_rate": 9.999935673501355e-07,
"loss": 0.257844477891922,
"step": 205,
"token_acc": 0.9278959810874704
},
{
"epoch": 0.043459915611814344,
"grad_norm": 0.71875,
"learning_rate": 9.99992681076094e-07,
"loss": 0.21238219738006592,
"step": 206,
"token_acc": 0.937776467118844
},
{
"epoch": 0.043670886075949364,
"grad_norm": 0.67578125,
"learning_rate": 9.999917376236578e-07,
"loss": 0.21476256847381592,
"step": 207,
"token_acc": 0.9357326478149101
},
{
"epoch": 0.04388185654008439,
"grad_norm": 0.97265625,
"learning_rate": 9.999907369929344e-07,
"loss": 0.24194155633449554,
"step": 208,
"token_acc": 0.9311714096624751
},
{
"epoch": 0.04409282700421941,
"grad_norm": 0.765625,
"learning_rate": 9.999896791840383e-07,
"loss": 0.2757856249809265,
"step": 209,
"token_acc": 0.9243792325056434
},
{
"epoch": 0.04430379746835443,
"grad_norm": 0.66015625,
"learning_rate": 9.999885641970906e-07,
"loss": 0.2318935990333557,
"step": 210,
"token_acc": 0.9357142857142857
},
{
"epoch": 0.04451476793248945,
"grad_norm": 0.64453125,
"learning_rate": 9.999873920322186e-07,
"loss": 0.2802179157733917,
"step": 211,
"token_acc": 0.9227716727716728
},
{
"epoch": 0.04472573839662447,
"grad_norm": 0.6796875,
"learning_rate": 9.999861626895565e-07,
"loss": 0.2558714747428894,
"step": 212,
"token_acc": 0.9255022321428571
},
{
"epoch": 0.04493670886075949,
"grad_norm": 0.890625,
"learning_rate": 9.99984876169245e-07,
"loss": 0.29882046580314636,
"step": 213,
"token_acc": 0.9213449414590213
},
{
"epoch": 0.04514767932489452,
"grad_norm": 0.76171875,
"learning_rate": 9.999835324714307e-07,
"loss": 0.24097202718257904,
"step": 214,
"token_acc": 0.9359388774610637
},
{
"epoch": 0.04535864978902954,
"grad_norm": 1.703125,
"learning_rate": 9.99982131596268e-07,
"loss": 0.28899186849594116,
"step": 215,
"token_acc": 0.9218701937865272
},
{
"epoch": 0.04556962025316456,
"grad_norm": 0.68359375,
"learning_rate": 9.999806735439165e-07,
"loss": 0.2710872292518616,
"step": 216,
"token_acc": 0.9248780487804878
},
{
"epoch": 0.04578059071729958,
"grad_norm": 0.6484375,
"learning_rate": 9.999791583145433e-07,
"loss": 0.2295331209897995,
"step": 217,
"token_acc": 0.9357304643261608
},
{
"epoch": 0.0459915611814346,
"grad_norm": 0.9921875,
"learning_rate": 9.999775859083216e-07,
"loss": 0.2171935886144638,
"step": 218,
"token_acc": 0.9356940509915014
},
{
"epoch": 0.046202531645569624,
"grad_norm": 0.94921875,
"learning_rate": 9.99975956325431e-07,
"loss": 0.2726757526397705,
"step": 219,
"token_acc": 0.9209691375829248
},
{
"epoch": 0.046413502109704644,
"grad_norm": 0.703125,
"learning_rate": 9.99974269566058e-07,
"loss": 0.27028149366378784,
"step": 220,
"token_acc": 0.9284016636957814
},
{
"epoch": 0.04662447257383966,
"grad_norm": 0.56640625,
"learning_rate": 9.999725256303957e-07,
"loss": 0.20975014567375183,
"step": 221,
"token_acc": 0.9344503233392122
},
{
"epoch": 0.04683544303797468,
"grad_norm": 0.80078125,
"learning_rate": 9.999707245186434e-07,
"loss": 0.3065168261528015,
"step": 222,
"token_acc": 0.9186879823594267
},
{
"epoch": 0.0470464135021097,
"grad_norm": 0.66796875,
"learning_rate": 9.999688662310072e-07,
"loss": 0.20764990150928497,
"step": 223,
"token_acc": 0.9452255418863503
},
{
"epoch": 0.04725738396624472,
"grad_norm": 0.89453125,
"learning_rate": 9.99966950767699e-07,
"loss": 0.2654411196708679,
"step": 224,
"token_acc": 0.9302244039270687
},
{
"epoch": 0.04746835443037975,
"grad_norm": 0.6953125,
"learning_rate": 9.999649781289385e-07,
"loss": 0.2514041066169739,
"step": 225,
"token_acc": 0.933082271147161
},
{
"epoch": 0.04767932489451477,
"grad_norm": 0.6328125,
"learning_rate": 9.99962948314951e-07,
"loss": 0.21037127077579498,
"step": 226,
"token_acc": 0.9351134846461949
},
{
"epoch": 0.04789029535864979,
"grad_norm": 0.59765625,
"learning_rate": 9.99960861325969e-07,
"loss": 0.21236909925937653,
"step": 227,
"token_acc": 0.940097449125824
},
{
"epoch": 0.04810126582278481,
"grad_norm": 0.6640625,
"learning_rate": 9.999587171622305e-07,
"loss": 0.21992863714694977,
"step": 228,
"token_acc": 0.9344711978055471
},
{
"epoch": 0.04831223628691983,
"grad_norm": 0.95703125,
"learning_rate": 9.999565158239812e-07,
"loss": 0.26401764154434204,
"step": 229,
"token_acc": 0.9244654262704805
},
{
"epoch": 0.04852320675105485,
"grad_norm": 0.73828125,
"learning_rate": 9.999542573114728e-07,
"loss": 0.24087585508823395,
"step": 230,
"token_acc": 0.926786751888437
},
{
"epoch": 0.048734177215189876,
"grad_norm": 0.703125,
"learning_rate": 9.999519416249634e-07,
"loss": 0.2533552646636963,
"step": 231,
"token_acc": 0.9275784028451342
},
{
"epoch": 0.048945147679324896,
"grad_norm": 1.4296875,
"learning_rate": 9.999495687647178e-07,
"loss": 0.2529897689819336,
"step": 232,
"token_acc": 0.9269195189639223
},
{
"epoch": 0.049156118143459916,
"grad_norm": 0.81640625,
"learning_rate": 9.999471387310077e-07,
"loss": 0.2788076400756836,
"step": 233,
"token_acc": 0.9202168861347793
},
{
"epoch": 0.049367088607594936,
"grad_norm": 0.68359375,
"learning_rate": 9.999446515241108e-07,
"loss": 0.2300492525100708,
"step": 234,
"token_acc": 0.9325668116842759
},
{
"epoch": 0.049578059071729956,
"grad_norm": 0.85546875,
"learning_rate": 9.999421071443115e-07,
"loss": 0.2711006700992584,
"step": 235,
"token_acc": 0.9220738900962434
},
{
"epoch": 0.049789029535864976,
"grad_norm": 0.66796875,
"learning_rate": 9.999395055919007e-07,
"loss": 0.24382656812667847,
"step": 236,
"token_acc": 0.9297777777777778
},
{
"epoch": 0.05,
"grad_norm": 0.76953125,
"learning_rate": 9.999368468671758e-07,
"loss": 0.2818126380443573,
"step": 237,
"token_acc": 0.9211531781868705
},
{
"epoch": 0.05021097046413502,
"grad_norm": 0.79296875,
"learning_rate": 9.999341309704413e-07,
"loss": 0.29420921206474304,
"step": 238,
"token_acc": 0.9187301587301587
},
{
"epoch": 0.05042194092827004,
"grad_norm": 0.6796875,
"learning_rate": 9.999313579020074e-07,
"loss": 0.24233081936836243,
"step": 239,
"token_acc": 0.9322516367776829
},
{
"epoch": 0.05063291139240506,
"grad_norm": 0.93359375,
"learning_rate": 9.999285276621913e-07,
"loss": 0.22199922800064087,
"step": 240,
"token_acc": 0.9347892956013534
},
{
"epoch": 0.05084388185654008,
"grad_norm": 0.6796875,
"learning_rate": 9.999256402513168e-07,
"loss": 0.2756049931049347,
"step": 241,
"token_acc": 0.9229754682141915
},
{
"epoch": 0.0510548523206751,
"grad_norm": 1.7421875,
"learning_rate": 9.999226956697138e-07,
"loss": 0.2459474354982376,
"step": 242,
"token_acc": 0.9287122207621551
},
{
"epoch": 0.05126582278481013,
"grad_norm": 0.7265625,
"learning_rate": 9.999196939177195e-07,
"loss": 0.26543667912483215,
"step": 243,
"token_acc": 0.9251445086705202
},
{
"epoch": 0.05147679324894515,
"grad_norm": 0.69921875,
"learning_rate": 9.999166349956768e-07,
"loss": 0.29306668043136597,
"step": 244,
"token_acc": 0.922089552238806
},
{
"epoch": 0.05168776371308017,
"grad_norm": 0.78515625,
"learning_rate": 9.999135189039356e-07,
"loss": 0.232993021607399,
"step": 245,
"token_acc": 0.933374460209747
},
{
"epoch": 0.05189873417721519,
"grad_norm": 0.703125,
"learning_rate": 9.999103456428522e-07,
"loss": 0.29452502727508545,
"step": 246,
"token_acc": 0.9255251432208784
},
{
"epoch": 0.05210970464135021,
"grad_norm": 0.75,
"learning_rate": 9.999071152127897e-07,
"loss": 0.2289431095123291,
"step": 247,
"token_acc": 0.9372047791053071
},
{
"epoch": 0.05232067510548523,
"grad_norm": 0.8046875,
"learning_rate": 9.999038276141175e-07,
"loss": 0.3194141983985901,
"step": 248,
"token_acc": 0.914375
},
{
"epoch": 0.052531645569620256,
"grad_norm": 0.68359375,
"learning_rate": 9.999004828472112e-07,
"loss": 0.24136003851890564,
"step": 249,
"token_acc": 0.9315025252525253
},
{
"epoch": 0.052742616033755275,
"grad_norm": 0.74609375,
"learning_rate": 9.998970809124537e-07,
"loss": 0.31663718819618225,
"step": 250,
"token_acc": 0.9186367823150138
},
{
"epoch": 0.052953586497890295,
"grad_norm": 0.765625,
"learning_rate": 9.998936218102338e-07,
"loss": 0.2638603448867798,
"step": 251,
"token_acc": 0.9242610837438424
},
{
"epoch": 0.053164556962025315,
"grad_norm": 0.62109375,
"learning_rate": 9.998901055409474e-07,
"loss": 0.26234734058380127,
"step": 252,
"token_acc": 0.9283480238839921
},
{
"epoch": 0.053375527426160335,
"grad_norm": 0.78125,
"learning_rate": 9.99886532104996e-07,
"loss": 0.27683377265930176,
"step": 253,
"token_acc": 0.9228208232445521
},
{
"epoch": 0.05358649789029536,
"grad_norm": 0.65234375,
"learning_rate": 9.99882901502789e-07,
"loss": 0.20958667993545532,
"step": 254,
"token_acc": 0.9367622259696459
},
{
"epoch": 0.05379746835443038,
"grad_norm": 0.765625,
"learning_rate": 9.998792137347412e-07,
"loss": 0.2769642174243927,
"step": 255,
"token_acc": 0.9259877573734001
},
{
"epoch": 0.0540084388185654,
"grad_norm": 0.7890625,
"learning_rate": 9.998754688012744e-07,
"loss": 0.291420578956604,
"step": 256,
"token_acc": 0.9195469067673541
},
{
"epoch": 0.05421940928270042,
"grad_norm": 0.7109375,
"learning_rate": 9.998716667028166e-07,
"loss": 0.2671175003051758,
"step": 257,
"token_acc": 0.9248520710059172
},
{
"epoch": 0.05443037974683544,
"grad_norm": 0.640625,
"learning_rate": 9.99867807439803e-07,
"loss": 0.2104148268699646,
"step": 258,
"token_acc": 0.9412735070933685
},
{
"epoch": 0.05464135021097046,
"grad_norm": 0.8125,
"learning_rate": 9.99863891012675e-07,
"loss": 0.25562331080436707,
"step": 259,
"token_acc": 0.9300422386483632
},
{
"epoch": 0.05485232067510549,
"grad_norm": 0.93359375,
"learning_rate": 9.998599174218797e-07,
"loss": 0.25945645570755005,
"step": 260,
"token_acc": 0.9278557114228457
},
{
"epoch": 0.05506329113924051,
"grad_norm": 0.76953125,
"learning_rate": 9.998558866678726e-07,
"loss": 0.2575325667858124,
"step": 261,
"token_acc": 0.9237336368810473
},
{
"epoch": 0.05527426160337553,
"grad_norm": 0.79296875,
"learning_rate": 9.998517987511139e-07,
"loss": 0.21312668919563293,
"step": 262,
"token_acc": 0.9391352244560727
},
{
"epoch": 0.05548523206751055,
"grad_norm": 1.5390625,
"learning_rate": 9.998476536720712e-07,
"loss": 0.27397406101226807,
"step": 263,
"token_acc": 0.9302109181141439
},
{
"epoch": 0.05569620253164557,
"grad_norm": 0.7265625,
"learning_rate": 9.998434514312187e-07,
"loss": 0.27095240354537964,
"step": 264,
"token_acc": 0.9266853059956508
},
{
"epoch": 0.05590717299578059,
"grad_norm": 0.76953125,
"learning_rate": 9.99839192029037e-07,
"loss": 0.237601175904274,
"step": 265,
"token_acc": 0.9364719228587635
},
{
"epoch": 0.056118143459915615,
"grad_norm": 0.7421875,
"learning_rate": 9.998348754660129e-07,
"loss": 0.2851409316062927,
"step": 266,
"token_acc": 0.9176470588235294
},
{
"epoch": 0.056329113924050635,
"grad_norm": 0.796875,
"learning_rate": 9.998305017426403e-07,
"loss": 0.26605701446533203,
"step": 267,
"token_acc": 0.9261460101867572
},
{
"epoch": 0.056540084388185655,
"grad_norm": 0.67578125,
"learning_rate": 9.998260708594192e-07,
"loss": 0.26237568259239197,
"step": 268,
"token_acc": 0.9257142857142857
},
{
"epoch": 0.056751054852320675,
"grad_norm": 0.859375,
"learning_rate": 9.998215828168566e-07,
"loss": 0.2315206527709961,
"step": 269,
"token_acc": 0.9332755632582322
},
{
"epoch": 0.056962025316455694,
"grad_norm": 0.70703125,
"learning_rate": 9.998170376154654e-07,
"loss": 0.26748204231262207,
"step": 270,
"token_acc": 0.9308067757680161
},
{
"epoch": 0.057172995780590714,
"grad_norm": 0.9609375,
"learning_rate": 9.998124352557655e-07,
"loss": 0.33397209644317627,
"step": 271,
"token_acc": 0.9161147902869757
},
{
"epoch": 0.05738396624472574,
"grad_norm": 0.71484375,
"learning_rate": 9.998077757382835e-07,
"loss": 0.2637864351272583,
"step": 272,
"token_acc": 0.9291949563530553
},
{
"epoch": 0.05759493670886076,
"grad_norm": 0.734375,
"learning_rate": 9.998030590635517e-07,
"loss": 0.2878430485725403,
"step": 273,
"token_acc": 0.919882100750268
},
{
"epoch": 0.05780590717299578,
"grad_norm": 0.65625,
"learning_rate": 9.997982852321099e-07,
"loss": 0.2438146024942398,
"step": 274,
"token_acc": 0.9312955692652832
},
{
"epoch": 0.0580168776371308,
"grad_norm": 0.74609375,
"learning_rate": 9.99793454244504e-07,
"loss": 0.2523839771747589,
"step": 275,
"token_acc": 0.9254349627174814
},
{
"epoch": 0.05822784810126582,
"grad_norm": 0.609375,
"learning_rate": 9.997885661012865e-07,
"loss": 0.23295487463474274,
"step": 276,
"token_acc": 0.937351934719663
},
{
"epoch": 0.05843881856540084,
"grad_norm": 0.77734375,
"learning_rate": 9.99783620803016e-07,
"loss": 0.287087619304657,
"step": 277,
"token_acc": 0.9255798969072165
},
{
"epoch": 0.05864978902953587,
"grad_norm": 0.65234375,
"learning_rate": 9.997786183502584e-07,
"loss": 0.23424138128757477,
"step": 278,
"token_acc": 0.924812030075188
},
{
"epoch": 0.05886075949367089,
"grad_norm": 0.9296875,
"learning_rate": 9.997735587435858e-07,
"loss": 0.25225332379341125,
"step": 279,
"token_acc": 0.9257325210327821
},
{
"epoch": 0.05907172995780591,
"grad_norm": 0.62890625,
"learning_rate": 9.997684419835767e-07,
"loss": 0.24867427349090576,
"step": 280,
"token_acc": 0.9290465631929047
},
{
"epoch": 0.05928270042194093,
"grad_norm": 1.265625,
"learning_rate": 9.997632680708163e-07,
"loss": 0.2555754482746124,
"step": 281,
"token_acc": 0.9300212056952438
},
{
"epoch": 0.05949367088607595,
"grad_norm": 0.6796875,
"learning_rate": 9.99758037005896e-07,
"loss": 0.25050830841064453,
"step": 282,
"token_acc": 0.9328039095907147
},
{
"epoch": 0.05970464135021097,
"grad_norm": 0.77734375,
"learning_rate": 9.997527487894144e-07,
"loss": 0.264704167842865,
"step": 283,
"token_acc": 0.9269878805793674
},
{
"epoch": 0.059915611814345994,
"grad_norm": 0.765625,
"learning_rate": 9.997474034219762e-07,
"loss": 0.29550492763519287,
"step": 284,
"token_acc": 0.9211438474870017
},
{
"epoch": 0.060126582278481014,
"grad_norm": 0.7109375,
"learning_rate": 9.997420009041927e-07,
"loss": 0.264403373003006,
"step": 285,
"token_acc": 0.9260048721071864
},
{
"epoch": 0.060337552742616034,
"grad_norm": 0.78125,
"learning_rate": 9.997365412366812e-07,
"loss": 0.2595897316932678,
"step": 286,
"token_acc": 0.9286173633440514
},
{
"epoch": 0.060548523206751054,
"grad_norm": 0.70703125,
"learning_rate": 9.997310244200667e-07,
"loss": 0.23976776003837585,
"step": 287,
"token_acc": 0.9318894271872328
},
{
"epoch": 0.060759493670886074,
"grad_norm": 0.9140625,
"learning_rate": 9.997254504549799e-07,
"loss": 0.26183438301086426,
"step": 288,
"token_acc": 0.9322949777495232
},
{
"epoch": 0.0609704641350211,
"grad_norm": 1.0078125,
"learning_rate": 9.99719819342058e-07,
"loss": 0.24600914120674133,
"step": 289,
"token_acc": 0.9311287236949987
},
{
"epoch": 0.06118143459915612,
"grad_norm": 0.83984375,
"learning_rate": 9.997141310819454e-07,
"loss": 0.3296029567718506,
"step": 290,
"token_acc": 0.9126184834123223
},
{
"epoch": 0.06139240506329114,
"grad_norm": 0.85546875,
"learning_rate": 9.997083856752923e-07,
"loss": 0.2794192433357239,
"step": 291,
"token_acc": 0.9190751445086706
},
{
"epoch": 0.06160337552742616,
"grad_norm": 0.6953125,
"learning_rate": 9.997025831227557e-07,
"loss": 0.23178298771381378,
"step": 292,
"token_acc": 0.9380645161290323
},
{
"epoch": 0.06181434599156118,
"grad_norm": 0.87109375,
"learning_rate": 9.996967234249994e-07,
"loss": 0.2989250123500824,
"step": 293,
"token_acc": 0.9201101928374655
},
{
"epoch": 0.0620253164556962,
"grad_norm": 0.625,
"learning_rate": 9.996908065826935e-07,
"loss": 0.20801636576652527,
"step": 294,
"token_acc": 0.9374828626268166
},
{
"epoch": 0.06223628691983123,
"grad_norm": 0.68359375,
"learning_rate": 9.996848325965142e-07,
"loss": 0.2513968050479889,
"step": 295,
"token_acc": 0.9286151960784313
},
{
"epoch": 0.06244725738396625,
"grad_norm": 0.77734375,
"learning_rate": 9.99678801467145e-07,
"loss": 0.23670442402362823,
"step": 296,
"token_acc": 0.9297945205479452
},
{
"epoch": 0.06265822784810127,
"grad_norm": 0.71875,
"learning_rate": 9.99672713195276e-07,
"loss": 0.3005760908126831,
"step": 297,
"token_acc": 0.9181008902077151
},
{
"epoch": 0.0628691983122363,
"grad_norm": 0.64453125,
"learning_rate": 9.996665677816027e-07,
"loss": 0.2198331356048584,
"step": 298,
"token_acc": 0.934411226357535
},
{
"epoch": 0.0630801687763713,
"grad_norm": 0.7421875,
"learning_rate": 9.996603652268283e-07,
"loss": 0.22930385172367096,
"step": 299,
"token_acc": 0.9343891402714932
},
{
"epoch": 0.06329113924050633,
"grad_norm": 0.6953125,
"learning_rate": 9.99654105531662e-07,
"loss": 0.26769182085990906,
"step": 300,
"token_acc": 0.9295731707317073
},
{
"epoch": 0.06350210970464135,
"grad_norm": 0.58203125,
"learning_rate": 9.9964778869682e-07,
"loss": 0.2113886922597885,
"step": 301,
"token_acc": 0.9383966244725739
},
{
"epoch": 0.06371308016877637,
"grad_norm": 0.78515625,
"learning_rate": 9.996414147230242e-07,
"loss": 0.2549387812614441,
"step": 302,
"token_acc": 0.9245056920311564
},
{
"epoch": 0.06392405063291139,
"grad_norm": 0.83203125,
"learning_rate": 9.996349836110035e-07,
"loss": 0.24877741932868958,
"step": 303,
"token_acc": 0.9278890600924499
},
{
"epoch": 0.06413502109704641,
"grad_norm": 0.85546875,
"learning_rate": 9.996284953614938e-07,
"loss": 0.2965357303619385,
"step": 304,
"token_acc": 0.9167351410572446
},
{
"epoch": 0.06434599156118144,
"grad_norm": 0.6484375,
"learning_rate": 9.996219499752365e-07,
"loss": 0.21444806456565857,
"step": 305,
"token_acc": 0.938101788170564
},
{
"epoch": 0.06455696202531645,
"grad_norm": 0.671875,
"learning_rate": 9.996153474529807e-07,
"loss": 0.24650560319423676,
"step": 306,
"token_acc": 0.928284854563691
},
{
"epoch": 0.06476793248945148,
"grad_norm": 0.73046875,
"learning_rate": 9.996086877954812e-07,
"loss": 0.26447594165802,
"step": 307,
"token_acc": 0.9272459499263623
},
{
"epoch": 0.06497890295358649,
"grad_norm": 0.6484375,
"learning_rate": 9.996019710034997e-07,
"loss": 0.22312676906585693,
"step": 308,
"token_acc": 0.9304399524375743
},
{
"epoch": 0.06518987341772152,
"grad_norm": 0.75,
"learning_rate": 9.99595197077804e-07,
"loss": 0.3124806582927704,
"step": 309,
"token_acc": 0.9134506242905789
},
{
"epoch": 0.06540084388185655,
"grad_norm": 0.6640625,
"learning_rate": 9.99588366019169e-07,
"loss": 0.21681739389896393,
"step": 310,
"token_acc": 0.9393859879296772
},
{
"epoch": 0.06561181434599156,
"grad_norm": 0.7734375,
"learning_rate": 9.99581477828376e-07,
"loss": 0.262542188167572,
"step": 311,
"token_acc": 0.9339063426200356
},
{
"epoch": 0.06582278481012659,
"grad_norm": 0.77734375,
"learning_rate": 9.995745325062126e-07,
"loss": 0.24062800407409668,
"step": 312,
"token_acc": 0.9331594391913922
},
{
"epoch": 0.0660337552742616,
"grad_norm": 0.71484375,
"learning_rate": 9.995675300534729e-07,
"loss": 0.26486438512802124,
"step": 313,
"token_acc": 0.9250824093497153
},
{
"epoch": 0.06624472573839663,
"grad_norm": 0.54296875,
"learning_rate": 9.995604704709578e-07,
"loss": 0.18465927243232727,
"step": 314,
"token_acc": 0.9464985994397759
},
{
"epoch": 0.06645569620253164,
"grad_norm": 0.91015625,
"learning_rate": 9.99553353759475e-07,
"loss": 0.2520803213119507,
"step": 315,
"token_acc": 0.9257213014119091
},
{
"epoch": 0.06666666666666667,
"grad_norm": 0.65625,
"learning_rate": 9.995461799198378e-07,
"loss": 0.29753151535987854,
"step": 316,
"token_acc": 0.9256516587677726
},
{
"epoch": 0.06687763713080169,
"grad_norm": 0.71875,
"learning_rate": 9.995389489528667e-07,
"loss": 0.2546613812446594,
"step": 317,
"token_acc": 0.9310240048617442
},
{
"epoch": 0.0670886075949367,
"grad_norm": 1.4140625,
"learning_rate": 9.995316608593886e-07,
"loss": 0.24808946251869202,
"step": 318,
"token_acc": 0.9316290130796671
},
{
"epoch": 0.06729957805907173,
"grad_norm": 0.73046875,
"learning_rate": 9.995243156402374e-07,
"loss": 0.2512444853782654,
"step": 319,
"token_acc": 0.9294703723125328
},
{
"epoch": 0.06751054852320675,
"grad_norm": 0.921875,
"learning_rate": 9.995169132962527e-07,
"loss": 0.2597760260105133,
"step": 320,
"token_acc": 0.9277310924369748
},
{
"epoch": 0.06772151898734177,
"grad_norm": 0.65234375,
"learning_rate": 9.99509453828281e-07,
"loss": 0.2552398443222046,
"step": 321,
"token_acc": 0.9278263321116437
},
{
"epoch": 0.0679324894514768,
"grad_norm": 0.75390625,
"learning_rate": 9.995019372371754e-07,
"loss": 0.29060834646224976,
"step": 322,
"token_acc": 0.9247988807275271
},
{
"epoch": 0.06814345991561181,
"grad_norm": 0.57421875,
"learning_rate": 9.994943635237955e-07,
"loss": 0.21358612179756165,
"step": 323,
"token_acc": 0.9360210341805434
},
{
"epoch": 0.06835443037974684,
"grad_norm": 0.8671875,
"learning_rate": 9.994867326890078e-07,
"loss": 0.2634425759315491,
"step": 324,
"token_acc": 0.9219944937289691
},
{
"epoch": 0.06856540084388185,
"grad_norm": 0.82421875,
"learning_rate": 9.994790447336842e-07,
"loss": 0.3185754120349884,
"step": 325,
"token_acc": 0.9133514986376022
},
{
"epoch": 0.06877637130801688,
"grad_norm": 0.73828125,
"learning_rate": 9.994712996587044e-07,
"loss": 0.29746031761169434,
"step": 326,
"token_acc": 0.9256002705444707
},
{
"epoch": 0.0689873417721519,
"grad_norm": 0.76953125,
"learning_rate": 9.994634974649541e-07,
"loss": 0.29588472843170166,
"step": 327,
"token_acc": 0.9272846380609236
},
{
"epoch": 0.06919831223628692,
"grad_norm": 0.875,
"learning_rate": 9.994556381533252e-07,
"loss": 0.277068167924881,
"step": 328,
"token_acc": 0.922756981580511
},
{
"epoch": 0.06940928270042195,
"grad_norm": 0.73046875,
"learning_rate": 9.994477217247168e-07,
"loss": 0.27129507064819336,
"step": 329,
"token_acc": 0.9270741068792442
},
{
"epoch": 0.06962025316455696,
"grad_norm": 0.8671875,
"learning_rate": 9.994397481800342e-07,
"loss": 0.24473360180854797,
"step": 330,
"token_acc": 0.9313361611876988
},
{
"epoch": 0.06983122362869199,
"grad_norm": 0.69140625,
"learning_rate": 9.994317175201893e-07,
"loss": 0.22818127274513245,
"step": 331,
"token_acc": 0.9380833851897946
},
{
"epoch": 0.070042194092827,
"grad_norm": 0.6953125,
"learning_rate": 9.994236297461003e-07,
"loss": 0.262783944606781,
"step": 332,
"token_acc": 0.9235968263297091
},
{
"epoch": 0.07025316455696203,
"grad_norm": 0.63671875,
"learning_rate": 9.994154848586919e-07,
"loss": 0.24861930310726166,
"step": 333,
"token_acc": 0.930621342992477
},
{
"epoch": 0.07046413502109705,
"grad_norm": 0.62109375,
"learning_rate": 9.99407282858896e-07,
"loss": 0.26165515184402466,
"step": 334,
"token_acc": 0.9247956403269755
},
{
"epoch": 0.07067510548523206,
"grad_norm": 0.64453125,
"learning_rate": 9.993990237476504e-07,
"loss": 0.23681169748306274,
"step": 335,
"token_acc": 0.9285078611687927
},
{
"epoch": 0.07088607594936709,
"grad_norm": 1.65625,
"learning_rate": 9.993907075258994e-07,
"loss": 0.2824210524559021,
"step": 336,
"token_acc": 0.925273390036452
},
{
"epoch": 0.0710970464135021,
"grad_norm": 0.6796875,
"learning_rate": 9.993823341945942e-07,
"loss": 0.2578677535057068,
"step": 337,
"token_acc": 0.9232230059685296
},
{
"epoch": 0.07130801687763713,
"grad_norm": 0.69140625,
"learning_rate": 9.993739037546924e-07,
"loss": 0.25358960032463074,
"step": 338,
"token_acc": 0.9312054539820267
},
{
"epoch": 0.07151898734177216,
"grad_norm": 0.83984375,
"learning_rate": 9.99365416207158e-07,
"loss": 0.2980523705482483,
"step": 339,
"token_acc": 0.9155807365439094
},
{
"epoch": 0.07172995780590717,
"grad_norm": 0.9296875,
"learning_rate": 9.993568715529616e-07,
"loss": 0.29448622465133667,
"step": 340,
"token_acc": 0.9224402207234825
},
{
"epoch": 0.0719409282700422,
"grad_norm": 0.6328125,
"learning_rate": 9.993482697930805e-07,
"loss": 0.2686302065849304,
"step": 341,
"token_acc": 0.9284134881149807
},
{
"epoch": 0.07215189873417721,
"grad_norm": 0.75,
"learning_rate": 9.993396109284985e-07,
"loss": 0.2800794839859009,
"step": 342,
"token_acc": 0.9271758436944938
},
{
"epoch": 0.07236286919831224,
"grad_norm": 0.765625,
"learning_rate": 9.993308949602054e-07,
"loss": 0.2576884329319,
"step": 343,
"token_acc": 0.9227618490345231
},
{
"epoch": 0.07257383966244725,
"grad_norm": 0.65625,
"learning_rate": 9.993221218891982e-07,
"loss": 0.24451857805252075,
"step": 344,
"token_acc": 0.933461117196057
},
{
"epoch": 0.07278481012658228,
"grad_norm": 0.95703125,
"learning_rate": 9.993132917164801e-07,
"loss": 0.2957763075828552,
"step": 345,
"token_acc": 0.9123943661971831
},
{
"epoch": 0.0729957805907173,
"grad_norm": 1.6875,
"learning_rate": 9.99304404443061e-07,
"loss": 0.253467321395874,
"step": 346,
"token_acc": 0.9318112633181126
},
{
"epoch": 0.07320675105485232,
"grad_norm": 0.8046875,
"learning_rate": 9.99295460069957e-07,
"loss": 0.2847754955291748,
"step": 347,
"token_acc": 0.9224688355123137
},
{
"epoch": 0.07341772151898734,
"grad_norm": 0.73046875,
"learning_rate": 9.992864585981913e-07,
"loss": 0.25176408886909485,
"step": 348,
"token_acc": 0.9317745035233824
},
{
"epoch": 0.07362869198312236,
"grad_norm": 0.96484375,
"learning_rate": 9.99277400028793e-07,
"loss": 0.282976359128952,
"step": 349,
"token_acc": 0.9220568335588634
},
{
"epoch": 0.07383966244725738,
"grad_norm": 0.64453125,
"learning_rate": 9.992682843627984e-07,
"loss": 0.2807369530200958,
"step": 350,
"token_acc": 0.9215148188803512
},
{
"epoch": 0.07405063291139241,
"grad_norm": 0.7265625,
"learning_rate": 9.992591116012495e-07,
"loss": 0.2882058322429657,
"step": 351,
"token_acc": 0.9216602528862012
},
{
"epoch": 0.07426160337552742,
"grad_norm": 0.73828125,
"learning_rate": 9.992498817451955e-07,
"loss": 0.27112358808517456,
"step": 352,
"token_acc": 0.9312214611872146
},
{
"epoch": 0.07447257383966245,
"grad_norm": 0.77734375,
"learning_rate": 9.99240594795692e-07,
"loss": 0.25564056634902954,
"step": 353,
"token_acc": 0.9308086560364465
},
{
"epoch": 0.07468354430379746,
"grad_norm": 0.66796875,
"learning_rate": 9.99231250753801e-07,
"loss": 0.21060852706432343,
"step": 354,
"token_acc": 0.9366489046773239
},
{
"epoch": 0.07489451476793249,
"grad_norm": 0.5546875,
"learning_rate": 9.992218496205908e-07,
"loss": 0.23291520774364471,
"step": 355,
"token_acc": 0.9379619852164731
},
{
"epoch": 0.0751054852320675,
"grad_norm": 0.703125,
"learning_rate": 9.99212391397137e-07,
"loss": 0.23014740645885468,
"step": 356,
"token_acc": 0.930849478390462
},
{
"epoch": 0.07531645569620253,
"grad_norm": 0.73046875,
"learning_rate": 9.992028760845207e-07,
"loss": 0.2653324604034424,
"step": 357,
"token_acc": 0.9264833574529667
},
{
"epoch": 0.07552742616033756,
"grad_norm": 0.76171875,
"learning_rate": 9.991933036838303e-07,
"loss": 0.23712849617004395,
"step": 358,
"token_acc": 0.9348139601961349
},
{
"epoch": 0.07573839662447257,
"grad_norm": 0.61328125,
"learning_rate": 9.991836741961605e-07,
"loss": 0.24832651019096375,
"step": 359,
"token_acc": 0.9297736506094022
},
{
"epoch": 0.0759493670886076,
"grad_norm": 2.21875,
"learning_rate": 9.991739876226127e-07,
"loss": 0.30170413851737976,
"step": 360,
"token_acc": 0.9175753688261706
},
{
"epoch": 0.07616033755274261,
"grad_norm": 0.6875,
"learning_rate": 9.991642439642944e-07,
"loss": 0.2096886932849884,
"step": 361,
"token_acc": 0.9416713404374649
},
{
"epoch": 0.07637130801687764,
"grad_norm": 0.609375,
"learning_rate": 9.991544432223198e-07,
"loss": 0.24230161309242249,
"step": 362,
"token_acc": 0.9317358595709111
},
{
"epoch": 0.07658227848101266,
"grad_norm": 0.79296875,
"learning_rate": 9.991445853978098e-07,
"loss": 0.2464846670627594,
"step": 363,
"token_acc": 0.9277708592777086
},
{
"epoch": 0.07679324894514768,
"grad_norm": 0.99609375,
"learning_rate": 9.991346704918918e-07,
"loss": 0.25032496452331543,
"step": 364,
"token_acc": 0.931261207411835
},
{
"epoch": 0.0770042194092827,
"grad_norm": 1.0,
"learning_rate": 9.991246985056995e-07,
"loss": 0.3197912871837616,
"step": 365,
"token_acc": 0.9187062937062938
},
{
"epoch": 0.07721518987341772,
"grad_norm": 0.6953125,
"learning_rate": 9.991146694403733e-07,
"loss": 0.2740510404109955,
"step": 366,
"token_acc": 0.9192671056398511
},
{
"epoch": 0.07742616033755274,
"grad_norm": 0.875,
"learning_rate": 9.991045832970603e-07,
"loss": 0.29503384232521057,
"step": 367,
"token_acc": 0.919302394324564
},
{
"epoch": 0.07763713080168777,
"grad_norm": 0.796875,
"learning_rate": 9.990944400769138e-07,
"loss": 0.27579015493392944,
"step": 368,
"token_acc": 0.9176502882239912
},
{
"epoch": 0.07784810126582278,
"grad_norm": 0.8828125,
"learning_rate": 9.99084239781094e-07,
"loss": 0.2694048583507538,
"step": 369,
"token_acc": 0.925148762918885
},
{
"epoch": 0.07805907172995781,
"grad_norm": 0.7578125,
"learning_rate": 9.990739824107669e-07,
"loss": 0.2885046601295471,
"step": 370,
"token_acc": 0.9219858156028369
},
{
"epoch": 0.07827004219409282,
"grad_norm": 1.109375,
"learning_rate": 9.99063667967106e-07,
"loss": 0.2373015433549881,
"step": 371,
"token_acc": 0.9304477611940298
},
{
"epoch": 0.07848101265822785,
"grad_norm": 0.93359375,
"learning_rate": 9.990532964512901e-07,
"loss": 0.29645416140556335,
"step": 372,
"token_acc": 0.918646080760095
},
{
"epoch": 0.07869198312236286,
"grad_norm": 0.66796875,
"learning_rate": 9.990428678645062e-07,
"loss": 0.24266409873962402,
"step": 373,
"token_acc": 0.9363528715216104
},
{
"epoch": 0.07890295358649789,
"grad_norm": 0.8046875,
"learning_rate": 9.990323822079464e-07,
"loss": 0.2219400256872177,
"step": 374,
"token_acc": 0.9366262814538676
},
{
"epoch": 0.07911392405063292,
"grad_norm": 0.609375,
"learning_rate": 9.9902183948281e-07,
"loss": 0.2171144187450409,
"step": 375,
"token_acc": 0.937206572769953
},
{
"epoch": 0.07932489451476793,
"grad_norm": 0.61328125,
"learning_rate": 9.990112396903027e-07,
"loss": 0.23284628987312317,
"step": 376,
"token_acc": 0.9356833642547928
},
{
"epoch": 0.07953586497890296,
"grad_norm": 0.765625,
"learning_rate": 9.990005828316363e-07,
"loss": 0.26610440015792847,
"step": 377,
"token_acc": 0.9276832460732984
},
{
"epoch": 0.07974683544303797,
"grad_norm": 0.55859375,
"learning_rate": 9.989898689080299e-07,
"loss": 0.19865182042121887,
"step": 378,
"token_acc": 0.946978672985782
},
{
"epoch": 0.079957805907173,
"grad_norm": 0.73046875,
"learning_rate": 9.989790979207085e-07,
"loss": 0.2547116279602051,
"step": 379,
"token_acc": 0.9301753306674869
},
{
"epoch": 0.08016877637130802,
"grad_norm": 0.71484375,
"learning_rate": 9.98968269870904e-07,
"loss": 0.2702917456626892,
"step": 380,
"token_acc": 0.9255110613273593
},
{
"epoch": 0.08037974683544304,
"grad_norm": 0.67578125,
"learning_rate": 9.989573847598545e-07,
"loss": 0.24545586109161377,
"step": 381,
"token_acc": 0.9353233830845771
},
{
"epoch": 0.08059071729957806,
"grad_norm": 0.70703125,
"learning_rate": 9.98946442588805e-07,
"loss": 0.25997835397720337,
"step": 382,
"token_acc": 0.9318181818181818
},
{
"epoch": 0.08080168776371308,
"grad_norm": 0.921875,
"learning_rate": 9.989354433590067e-07,
"loss": 0.2865683436393738,
"step": 383,
"token_acc": 0.9229352164568622
},
{
"epoch": 0.0810126582278481,
"grad_norm": 0.93359375,
"learning_rate": 9.989243870717174e-07,
"loss": 0.25773969292640686,
"step": 384,
"token_acc": 0.9284097340124505
},
{
"epoch": 0.08122362869198312,
"grad_norm": 0.87890625,
"learning_rate": 9.989132737282015e-07,
"loss": 0.2665586471557617,
"step": 385,
"token_acc": 0.92599672310213
},
{
"epoch": 0.08143459915611814,
"grad_norm": 0.7890625,
"learning_rate": 9.989021033297302e-07,
"loss": 0.29251331090927124,
"step": 386,
"token_acc": 0.9255349500713267
},
{
"epoch": 0.08164556962025317,
"grad_norm": 0.90625,
"learning_rate": 9.988908758775807e-07,
"loss": 0.31350889801979065,
"step": 387,
"token_acc": 0.9161179501860864
},
{
"epoch": 0.08185654008438818,
"grad_norm": 0.77734375,
"learning_rate": 9.98879591373037e-07,
"loss": 0.2779198884963989,
"step": 388,
"token_acc": 0.9238483234095894
},
{
"epoch": 0.08206751054852321,
"grad_norm": 0.60546875,
"learning_rate": 9.988682498173895e-07,
"loss": 0.22718225419521332,
"step": 389,
"token_acc": 0.9420247204237787
},
{
"epoch": 0.08227848101265822,
"grad_norm": 0.609375,
"learning_rate": 9.98856851211935e-07,
"loss": 0.22414159774780273,
"step": 390,
"token_acc": 0.9357366771159875
},
{
"epoch": 0.08248945147679325,
"grad_norm": 0.74609375,
"learning_rate": 9.988453955579776e-07,
"loss": 0.2700081467628479,
"step": 391,
"token_acc": 0.9235555555555556
},
{
"epoch": 0.08270042194092828,
"grad_norm": 0.66796875,
"learning_rate": 9.98833882856827e-07,
"loss": 0.24140852689743042,
"step": 392,
"token_acc": 0.9277988101676582
},
{
"epoch": 0.08291139240506329,
"grad_norm": 0.8359375,
"learning_rate": 9.988223131097996e-07,
"loss": 0.28851526975631714,
"step": 393,
"token_acc": 0.9173528514791095
},
{
"epoch": 0.08312236286919832,
"grad_norm": 0.6796875,
"learning_rate": 9.98810686318219e-07,
"loss": 0.2673947215080261,
"step": 394,
"token_acc": 0.9220706930141943
},
{
"epoch": 0.08333333333333333,
"grad_norm": 0.66796875,
"learning_rate": 9.98799002483414e-07,
"loss": 0.27388495206832886,
"step": 395,
"token_acc": 0.9260257562144355
},
{
"epoch": 0.08354430379746836,
"grad_norm": 0.6484375,
"learning_rate": 9.987872616067216e-07,
"loss": 0.2556672692298889,
"step": 396,
"token_acc": 0.9255903349807798
},
{
"epoch": 0.08375527426160338,
"grad_norm": 0.96875,
"learning_rate": 9.987754636894843e-07,
"loss": 0.32614314556121826,
"step": 397,
"token_acc": 0.9155054847316929
},
{
"epoch": 0.0839662447257384,
"grad_norm": 0.60546875,
"learning_rate": 9.987636087330509e-07,
"loss": 0.23008616268634796,
"step": 398,
"token_acc": 0.9367875647668393
},
{
"epoch": 0.08417721518987342,
"grad_norm": 0.73828125,
"learning_rate": 9.987516967387775e-07,
"loss": 0.2754250764846802,
"step": 399,
"token_acc": 0.9202546998180715
},
{
"epoch": 0.08438818565400844,
"grad_norm": 0.671875,
"learning_rate": 9.98739727708026e-07,
"loss": 0.23332414031028748,
"step": 400,
"token_acc": 0.9356028368794326
},
{
"epoch": 0.08438818565400844,
"eval_loss": 0.43364420533180237,
"eval_runtime": 245.8014,
"eval_samples_per_second": 137.123,
"eval_steps_per_second": 2.144,
"eval_token_acc": 0.8990711657542853,
"step": 400
},
{
"epoch": 0.08459915611814346,
"grad_norm": 0.7578125,
"learning_rate": 9.987277016421654e-07,
"loss": 0.2699899673461914,
"step": 401,
"token_acc": 0.9288135593220339
},
{
"epoch": 0.08481012658227848,
"grad_norm": 0.78515625,
"learning_rate": 9.98715618542571e-07,
"loss": 0.25560492277145386,
"step": 402,
"token_acc": 0.9252772913018097
},
{
"epoch": 0.0850210970464135,
"grad_norm": 0.7890625,
"learning_rate": 9.987034784106244e-07,
"loss": 0.3024590015411377,
"step": 403,
"token_acc": 0.9186206896551724
},
{
"epoch": 0.08523206751054853,
"grad_norm": 0.703125,
"learning_rate": 9.98691281247714e-07,
"loss": 0.2880774438381195,
"step": 404,
"token_acc": 0.9188865609099072
},
{
"epoch": 0.08544303797468354,
"grad_norm": 0.859375,
"learning_rate": 9.986790270552347e-07,
"loss": 0.2641194760799408,
"step": 405,
"token_acc": 0.9306480920654149
},
{
"epoch": 0.08565400843881857,
"grad_norm": 0.72265625,
"learning_rate": 9.98666715834588e-07,
"loss": 0.25727787613868713,
"step": 406,
"token_acc": 0.9288548752834467
},
{
"epoch": 0.08586497890295358,
"grad_norm": 0.61328125,
"learning_rate": 9.986543475871818e-07,
"loss": 0.2398534119129181,
"step": 407,
"token_acc": 0.9377962085308057
},
{
"epoch": 0.08607594936708861,
"grad_norm": 0.671875,
"learning_rate": 9.986419223144302e-07,
"loss": 0.25430333614349365,
"step": 408,
"token_acc": 0.9305912596401028
},
{
"epoch": 0.08628691983122364,
"grad_norm": 0.60546875,
"learning_rate": 9.986294400177544e-07,
"loss": 0.20513233542442322,
"step": 409,
"token_acc": 0.9459538416593631
},
{
"epoch": 0.08649789029535865,
"grad_norm": 0.5625,
"learning_rate": 9.986169006985817e-07,
"loss": 0.20912200212478638,
"step": 410,
"token_acc": 0.9390818128310771
},
{
"epoch": 0.08670886075949367,
"grad_norm": 0.67578125,
"learning_rate": 9.986043043583462e-07,
"loss": 0.2466573864221573,
"step": 411,
"token_acc": 0.9298196948682386
},
{
"epoch": 0.08691983122362869,
"grad_norm": 0.7578125,
"learning_rate": 9.98591650998488e-07,
"loss": 0.23664775490760803,
"step": 412,
"token_acc": 0.9329004329004329
},
{
"epoch": 0.08713080168776371,
"grad_norm": 0.640625,
"learning_rate": 9.985789406204547e-07,
"loss": 0.23415768146514893,
"step": 413,
"token_acc": 0.9317912218268091
},
{
"epoch": 0.08734177215189873,
"grad_norm": 0.75,
"learning_rate": 9.985661732256998e-07,
"loss": 0.2954852283000946,
"step": 414,
"token_acc": 0.9218203033838973
},
{
"epoch": 0.08755274261603375,
"grad_norm": 0.703125,
"learning_rate": 9.98553348815683e-07,
"loss": 0.2565650939941406,
"step": 415,
"token_acc": 0.9298298906439855
},
{
"epoch": 0.08776371308016878,
"grad_norm": 0.7578125,
"learning_rate": 9.98540467391871e-07,
"loss": 0.22425369918346405,
"step": 416,
"token_acc": 0.9359098228663446
},
{
"epoch": 0.0879746835443038,
"grad_norm": 0.73046875,
"learning_rate": 9.98527528955737e-07,
"loss": 0.2637927532196045,
"step": 417,
"token_acc": 0.9261443414771132
},
{
"epoch": 0.08818565400843882,
"grad_norm": 0.70703125,
"learning_rate": 9.985145335087605e-07,
"loss": 0.27013063430786133,
"step": 418,
"token_acc": 0.9248719408081958
},
{
"epoch": 0.08839662447257383,
"grad_norm": 0.62890625,
"learning_rate": 9.985014810524278e-07,
"loss": 0.25381606817245483,
"step": 419,
"token_acc": 0.9342265529841657
},
{
"epoch": 0.08860759493670886,
"grad_norm": 1.59375,
"learning_rate": 9.984883715882315e-07,
"loss": 0.2093265801668167,
"step": 420,
"token_acc": 0.936447410231967
},
{
"epoch": 0.08881856540084389,
"grad_norm": 0.65625,
"learning_rate": 9.984752051176707e-07,
"loss": 0.2633010447025299,
"step": 421,
"token_acc": 0.9257308401369502
},
{
"epoch": 0.0890295358649789,
"grad_norm": 0.6796875,
"learning_rate": 9.98461981642251e-07,
"loss": 0.28037169575691223,
"step": 422,
"token_acc": 0.9264617239300783
},
{
"epoch": 0.08924050632911393,
"grad_norm": 0.640625,
"learning_rate": 9.984487011634848e-07,
"loss": 0.23874756693840027,
"step": 423,
"token_acc": 0.9345043167609407
},
{
"epoch": 0.08945147679324894,
"grad_norm": 4.9375,
"learning_rate": 9.984353636828908e-07,
"loss": 0.2935020923614502,
"step": 424,
"token_acc": 0.9228658536585366
},
{
"epoch": 0.08966244725738397,
"grad_norm": 0.62890625,
"learning_rate": 9.984219692019943e-07,
"loss": 0.2578403949737549,
"step": 425,
"token_acc": 0.9282193468884782
},
{
"epoch": 0.08987341772151898,
"grad_norm": 0.73828125,
"learning_rate": 9.98408517722327e-07,
"loss": 0.24426788091659546,
"step": 426,
"token_acc": 0.9371653987038603
},
{
"epoch": 0.09008438818565401,
"grad_norm": 0.80859375,
"learning_rate": 9.983950092454272e-07,
"loss": 0.2677040994167328,
"step": 427,
"token_acc": 0.9215219976218787
},
{
"epoch": 0.09029535864978903,
"grad_norm": 0.73828125,
"learning_rate": 9.983814437728396e-07,
"loss": 0.2604065239429474,
"step": 428,
"token_acc": 0.9289383561643836
},
{
"epoch": 0.09050632911392405,
"grad_norm": 1.3125,
"learning_rate": 9.983678213061157e-07,
"loss": 0.24889585375785828,
"step": 429,
"token_acc": 0.9251644736842105
},
{
"epoch": 0.09071729957805907,
"grad_norm": 0.796875,
"learning_rate": 9.983541418468134e-07,
"loss": 0.2805905342102051,
"step": 430,
"token_acc": 0.9248591108328115
},
{
"epoch": 0.09092827004219409,
"grad_norm": 0.7734375,
"learning_rate": 9.983404053964967e-07,
"loss": 0.2725668251514435,
"step": 431,
"token_acc": 0.9280293116985082
},
{
"epoch": 0.09113924050632911,
"grad_norm": 0.84375,
"learning_rate": 9.98326611956737e-07,
"loss": 0.3275222182273865,
"step": 432,
"token_acc": 0.9095477386934674
},
{
"epoch": 0.09135021097046414,
"grad_norm": 0.91015625,
"learning_rate": 9.98312761529111e-07,
"loss": 0.27813225984573364,
"step": 433,
"token_acc": 0.9216018048505358
},
{
"epoch": 0.09156118143459915,
"grad_norm": 0.74609375,
"learning_rate": 9.982988541152036e-07,
"loss": 0.2637915015220642,
"step": 434,
"token_acc": 0.9244929797191888
},
{
"epoch": 0.09177215189873418,
"grad_norm": 1.0703125,
"learning_rate": 9.982848897166042e-07,
"loss": 0.2686794102191925,
"step": 435,
"token_acc": 0.9231661351116266
},
{
"epoch": 0.0919831223628692,
"grad_norm": 0.62890625,
"learning_rate": 9.982708683349105e-07,
"loss": 0.24819687008857727,
"step": 436,
"token_acc": 0.9376739009460211
},
{
"epoch": 0.09219409282700422,
"grad_norm": 0.75,
"learning_rate": 9.982567899717256e-07,
"loss": 0.25789859890937805,
"step": 437,
"token_acc": 0.9243027888446215
},
{
"epoch": 0.09240506329113925,
"grad_norm": 0.65625,
"learning_rate": 9.982426546286596e-07,
"loss": 0.2573246359825134,
"step": 438,
"token_acc": 0.9290590679726922
},
{
"epoch": 0.09261603375527426,
"grad_norm": 0.7578125,
"learning_rate": 9.98228462307329e-07,
"loss": 0.2979215383529663,
"step": 439,
"token_acc": 0.9225908372827805
},
{
"epoch": 0.09282700421940929,
"grad_norm": 0.703125,
"learning_rate": 9.982142130093566e-07,
"loss": 0.2403128445148468,
"step": 440,
"token_acc": 0.9290377519159807
},
{
"epoch": 0.0930379746835443,
"grad_norm": 0.7890625,
"learning_rate": 9.98199906736372e-07,
"loss": 0.2767883241176605,
"step": 441,
"token_acc": 0.9238754325259516
},
{
"epoch": 0.09324894514767933,
"grad_norm": 0.609375,
"learning_rate": 9.981855434900115e-07,
"loss": 0.25662270188331604,
"step": 442,
"token_acc": 0.9294367050272562
},
{
"epoch": 0.09345991561181434,
"grad_norm": 0.7109375,
"learning_rate": 9.981711232719175e-07,
"loss": 0.24665901064872742,
"step": 443,
"token_acc": 0.9237830319888735
},
{
"epoch": 0.09367088607594937,
"grad_norm": 0.7890625,
"learning_rate": 9.98156646083739e-07,
"loss": 0.23571115732192993,
"step": 444,
"token_acc": 0.9281145293938471
},
{
"epoch": 0.0938818565400844,
"grad_norm": 1.21875,
"learning_rate": 9.981421119271316e-07,
"loss": 0.2607622742652893,
"step": 445,
"token_acc": 0.9253941441441441
},
{
"epoch": 0.0940928270042194,
"grad_norm": 0.73046875,
"learning_rate": 9.981275208037575e-07,
"loss": 0.2898206114768982,
"step": 446,
"token_acc": 0.918967587034814
},
{
"epoch": 0.09430379746835443,
"grad_norm": 0.6640625,
"learning_rate": 9.981128727152854e-07,
"loss": 0.2372782677412033,
"step": 447,
"token_acc": 0.9295408605255558
},
{
"epoch": 0.09451476793248945,
"grad_norm": 0.65234375,
"learning_rate": 9.980981676633903e-07,
"loss": 0.22987963259220123,
"step": 448,
"token_acc": 0.9354383986467437
},
{
"epoch": 0.09472573839662447,
"grad_norm": 0.95703125,
"learning_rate": 9.980834056497538e-07,
"loss": 0.26702481508255005,
"step": 449,
"token_acc": 0.9252548131370328
},
{
"epoch": 0.0949367088607595,
"grad_norm": 0.76171875,
"learning_rate": 9.98068586676064e-07,
"loss": 0.27268826961517334,
"step": 450,
"token_acc": 0.9244391971664699
},
{
"epoch": 0.09514767932489451,
"grad_norm": 0.93359375,
"learning_rate": 9.98053710744016e-07,
"loss": 0.22254578769207,
"step": 451,
"token_acc": 0.9397944199706314
},
{
"epoch": 0.09535864978902954,
"grad_norm": 0.6484375,
"learning_rate": 9.980387778553103e-07,
"loss": 0.2529526948928833,
"step": 452,
"token_acc": 0.9301044083526682
},
{
"epoch": 0.09556962025316455,
"grad_norm": 0.93359375,
"learning_rate": 9.980237880116553e-07,
"loss": 0.2600526809692383,
"step": 453,
"token_acc": 0.9255893212155638
},
{
"epoch": 0.09578059071729958,
"grad_norm": 0.74609375,
"learning_rate": 9.980087412147648e-07,
"loss": 0.2552299499511719,
"step": 454,
"token_acc": 0.9276672694394213
},
{
"epoch": 0.09599156118143459,
"grad_norm": 0.984375,
"learning_rate": 9.979936374663595e-07,
"loss": 0.28409841656684875,
"step": 455,
"token_acc": 0.9230544177881802
},
{
"epoch": 0.09620253164556962,
"grad_norm": 0.83984375,
"learning_rate": 9.979784767681668e-07,
"loss": 0.256397545337677,
"step": 456,
"token_acc": 0.9331357048748353
},
{
"epoch": 0.09641350210970465,
"grad_norm": 0.69140625,
"learning_rate": 9.979632591219207e-07,
"loss": 0.2313995659351349,
"step": 457,
"token_acc": 0.9336188436830836
},
{
"epoch": 0.09662447257383966,
"grad_norm": 0.87890625,
"learning_rate": 9.97947984529361e-07,
"loss": 0.2967644929885864,
"step": 458,
"token_acc": 0.9195630585898709
},
{
"epoch": 0.09683544303797469,
"grad_norm": 0.76953125,
"learning_rate": 9.979326529922348e-07,
"loss": 0.30269140005111694,
"step": 459,
"token_acc": 0.9189985272459499
},
{
"epoch": 0.0970464135021097,
"grad_norm": 0.68359375,
"learning_rate": 9.97917264512295e-07,
"loss": 0.2591363787651062,
"step": 460,
"token_acc": 0.9290578887627696
},
{
"epoch": 0.09725738396624473,
"grad_norm": 0.84375,
"learning_rate": 9.979018190913018e-07,
"loss": 0.32560282945632935,
"step": 461,
"token_acc": 0.9178757980266976
},
{
"epoch": 0.09746835443037975,
"grad_norm": 0.83984375,
"learning_rate": 9.978863167310213e-07,
"loss": 0.2893942892551422,
"step": 462,
"token_acc": 0.924191063174114
},
{
"epoch": 0.09767932489451477,
"grad_norm": 0.625,
"learning_rate": 9.978707574332266e-07,
"loss": 0.2492993026971817,
"step": 463,
"token_acc": 0.9310970081595649
},
{
"epoch": 0.09789029535864979,
"grad_norm": 0.734375,
"learning_rate": 9.978551411996967e-07,
"loss": 0.27646076679229736,
"step": 464,
"token_acc": 0.9283480238839921
},
{
"epoch": 0.0981012658227848,
"grad_norm": 0.87890625,
"learning_rate": 9.978394680322176e-07,
"loss": 0.22209137678146362,
"step": 465,
"token_acc": 0.9360902255639098
},
{
"epoch": 0.09831223628691983,
"grad_norm": 0.84375,
"learning_rate": 9.978237379325818e-07,
"loss": 0.2588399648666382,
"step": 466,
"token_acc": 0.9257776408992916
},
{
"epoch": 0.09852320675105486,
"grad_norm": 0.88671875,
"learning_rate": 9.978079509025878e-07,
"loss": 0.3038383722305298,
"step": 467,
"token_acc": 0.9133605600933489
},
{
"epoch": 0.09873417721518987,
"grad_norm": 0.8125,
"learning_rate": 9.977921069440415e-07,
"loss": 0.24923110008239746,
"step": 468,
"token_acc": 0.9317915690866511
},
{
"epoch": 0.0989451476793249,
"grad_norm": 0.8515625,
"learning_rate": 9.97776206058754e-07,
"loss": 0.23833706974983215,
"step": 469,
"token_acc": 0.9329593267882188
},
{
"epoch": 0.09915611814345991,
"grad_norm": 0.66015625,
"learning_rate": 9.977602482485445e-07,
"loss": 0.25747478008270264,
"step": 470,
"token_acc": 0.9295946357817738
},
{
"epoch": 0.09936708860759494,
"grad_norm": 0.68359375,
"learning_rate": 9.977442335152377e-07,
"loss": 0.2688140571117401,
"step": 471,
"token_acc": 0.9248041775456919
},
{
"epoch": 0.09957805907172995,
"grad_norm": 0.65625,
"learning_rate": 9.977281618606649e-07,
"loss": 0.19290462136268616,
"step": 472,
"token_acc": 0.9412288512911843
},
{
"epoch": 0.09978902953586498,
"grad_norm": 0.66015625,
"learning_rate": 9.977120332866638e-07,
"loss": 0.24847334623336792,
"step": 473,
"token_acc": 0.9335578689528475
},
{
"epoch": 0.1,
"grad_norm": 0.73828125,
"learning_rate": 9.976958477950794e-07,
"loss": 0.24599069356918335,
"step": 474,
"token_acc": 0.9284507042253521
},
{
"epoch": 0.10021097046413502,
"grad_norm": 0.83203125,
"learning_rate": 9.976796053877622e-07,
"loss": 0.2468043714761734,
"step": 475,
"token_acc": 0.9286099137931034
},
{
"epoch": 0.10042194092827005,
"grad_norm": 0.73828125,
"learning_rate": 9.976633060665697e-07,
"loss": 0.2741178572177887,
"step": 476,
"token_acc": 0.9224250325945241
},
{
"epoch": 0.10063291139240506,
"grad_norm": 0.6484375,
"learning_rate": 9.97646949833366e-07,
"loss": 0.23784510791301727,
"step": 477,
"token_acc": 0.9328483491885842
},
{
"epoch": 0.10084388185654009,
"grad_norm": 0.66015625,
"learning_rate": 9.976305366900216e-07,
"loss": 0.23838309943675995,
"step": 478,
"token_acc": 0.9320939839917377
},
{
"epoch": 0.10105485232067511,
"grad_norm": 0.86328125,
"learning_rate": 9.976140666384134e-07,
"loss": 0.2787632346153259,
"step": 479,
"token_acc": 0.9210836277974087
},
{
"epoch": 0.10126582278481013,
"grad_norm": 0.7265625,
"learning_rate": 9.97597539680425e-07,
"loss": 0.2557927370071411,
"step": 480,
"token_acc": 0.9304747320061256
},
{
"epoch": 0.10147679324894515,
"grad_norm": 0.8359375,
"learning_rate": 9.975809558179463e-07,
"loss": 0.2617788314819336,
"step": 481,
"token_acc": 0.9297163995067818
},
{
"epoch": 0.10168776371308016,
"grad_norm": 0.77734375,
"learning_rate": 9.975643150528737e-07,
"loss": 0.24790287017822266,
"step": 482,
"token_acc": 0.9321890827236916
},
{
"epoch": 0.10189873417721519,
"grad_norm": 0.62109375,
"learning_rate": 9.975476173871102e-07,
"loss": 0.22530625760555267,
"step": 483,
"token_acc": 0.9375520399666945
},
{
"epoch": 0.1021097046413502,
"grad_norm": 0.68359375,
"learning_rate": 9.975308628225657e-07,
"loss": 0.23794007301330566,
"step": 484,
"token_acc": 0.9279202279202279
},
{
"epoch": 0.10232067510548523,
"grad_norm": 0.71875,
"learning_rate": 9.975140513611558e-07,
"loss": 0.2270554155111313,
"step": 485,
"token_acc": 0.9365750528541226
},
{
"epoch": 0.10253164556962026,
"grad_norm": 0.77734375,
"learning_rate": 9.974971830048033e-07,
"loss": 0.23995614051818848,
"step": 486,
"token_acc": 0.9316101238556812
},
{
"epoch": 0.10274261603375527,
"grad_norm": 0.58203125,
"learning_rate": 9.974802577554372e-07,
"loss": 0.2599806487560272,
"step": 487,
"token_acc": 0.9297071129707113
},
{
"epoch": 0.1029535864978903,
"grad_norm": 0.703125,
"learning_rate": 9.974632756149928e-07,
"loss": 0.2610231935977936,
"step": 488,
"token_acc": 0.9277639922801213
},
{
"epoch": 0.10316455696202531,
"grad_norm": 0.69921875,
"learning_rate": 9.974462365854124e-07,
"loss": 0.2433297038078308,
"step": 489,
"token_acc": 0.9279082468596396
},
{
"epoch": 0.10337552742616034,
"grad_norm": 0.765625,
"learning_rate": 9.974291406686446e-07,
"loss": 0.21656793355941772,
"step": 490,
"token_acc": 0.9369074861065708
},
{
"epoch": 0.10358649789029536,
"grad_norm": 0.77734375,
"learning_rate": 9.974119878666442e-07,
"loss": 0.2721899151802063,
"step": 491,
"token_acc": 0.9287462605384824
},
{
"epoch": 0.10379746835443038,
"grad_norm": 0.87890625,
"learning_rate": 9.973947781813731e-07,
"loss": 0.25939926505088806,
"step": 492,
"token_acc": 0.9284844796104686
},
{
"epoch": 0.1040084388185654,
"grad_norm": 0.73828125,
"learning_rate": 9.973775116147992e-07,
"loss": 0.2712750732898712,
"step": 493,
"token_acc": 0.9242995689655172
},
{
"epoch": 0.10421940928270042,
"grad_norm": 0.8203125,
"learning_rate": 9.97360188168897e-07,
"loss": 0.2513953447341919,
"step": 494,
"token_acc": 0.929803328290469
},
{
"epoch": 0.10443037974683544,
"grad_norm": 0.62890625,
"learning_rate": 9.973428078456475e-07,
"loss": 0.2344273030757904,
"step": 495,
"token_acc": 0.9309220278683664
},
{
"epoch": 0.10464135021097046,
"grad_norm": 0.6796875,
"learning_rate": 9.973253706470388e-07,
"loss": 0.24709591269493103,
"step": 496,
"token_acc": 0.9282845668387837
},
{
"epoch": 0.10485232067510548,
"grad_norm": 0.734375,
"learning_rate": 9.973078765750644e-07,
"loss": 0.26154980063438416,
"step": 497,
"token_acc": 0.9249655172413793
},
{
"epoch": 0.10506329113924051,
"grad_norm": 0.671875,
"learning_rate": 9.972903256317251e-07,
"loss": 0.2260134369134903,
"step": 498,
"token_acc": 0.9395458566794456
},
{
"epoch": 0.10527426160337552,
"grad_norm": 0.80859375,
"learning_rate": 9.972727178190281e-07,
"loss": 0.33081650733947754,
"step": 499,
"token_acc": 0.9097568121886903
},
{
"epoch": 0.10548523206751055,
"grad_norm": 0.7109375,
"learning_rate": 9.97255053138987e-07,
"loss": 0.23815643787384033,
"step": 500,
"token_acc": 0.929639889196676
},
{
"epoch": 0.10569620253164556,
"grad_norm": 0.73046875,
"learning_rate": 9.972373315936218e-07,
"loss": 0.2648988962173462,
"step": 501,
"token_acc": 0.9245283018867925
},
{
"epoch": 0.10590717299578059,
"grad_norm": 0.765625,
"learning_rate": 9.972195531849592e-07,
"loss": 0.2421625256538391,
"step": 502,
"token_acc": 0.9339531901250401
},
{
"epoch": 0.10611814345991562,
"grad_norm": 0.53125,
"learning_rate": 9.97201717915032e-07,
"loss": 0.2174941599369049,
"step": 503,
"token_acc": 0.9379822806516147
},
{
"epoch": 0.10632911392405063,
"grad_norm": 0.9375,
"learning_rate": 9.971838257858804e-07,
"loss": 0.24187928438186646,
"step": 504,
"token_acc": 0.924503742271396
},
{
"epoch": 0.10654008438818566,
"grad_norm": 0.6640625,
"learning_rate": 9.9716587679955e-07,
"loss": 0.2514258921146393,
"step": 505,
"token_acc": 0.9230990783410138
},
{
"epoch": 0.10675105485232067,
"grad_norm": 0.69921875,
"learning_rate": 9.971478709580937e-07,
"loss": 0.282311350107193,
"step": 506,
"token_acc": 0.9201725997842503
},
{
"epoch": 0.1069620253164557,
"grad_norm": 0.66015625,
"learning_rate": 9.971298082635705e-07,
"loss": 0.2298332005739212,
"step": 507,
"token_acc": 0.936689779921616
},
{
"epoch": 0.10717299578059072,
"grad_norm": 0.7109375,
"learning_rate": 9.971116887180461e-07,
"loss": 0.26396387815475464,
"step": 508,
"token_acc": 0.9267986176562991
},
{
"epoch": 0.10738396624472574,
"grad_norm": 1.3046875,
"learning_rate": 9.970935123235926e-07,
"loss": 0.2639835476875305,
"step": 509,
"token_acc": 0.9236704326260677
},
{
"epoch": 0.10759493670886076,
"grad_norm": 0.75390625,
"learning_rate": 9.970752790822886e-07,
"loss": 0.27394697070121765,
"step": 510,
"token_acc": 0.9261565836298933
},
{
"epoch": 0.10780590717299578,
"grad_norm": 0.73828125,
"learning_rate": 9.97056988996219e-07,
"loss": 0.2229761779308319,
"step": 511,
"token_acc": 0.9351882160392798
},
{
"epoch": 0.1080168776371308,
"grad_norm": 0.73828125,
"learning_rate": 9.970386420674758e-07,
"loss": 0.26045358180999756,
"step": 512,
"token_acc": 0.9279547484370348
},
{
"epoch": 0.10822784810126582,
"grad_norm": 0.66015625,
"learning_rate": 9.97020238298157e-07,
"loss": 0.23026743531227112,
"step": 513,
"token_acc": 0.9363579080025205
},
{
"epoch": 0.10843881856540084,
"grad_norm": 0.76171875,
"learning_rate": 9.970017776903671e-07,
"loss": 0.2587951421737671,
"step": 514,
"token_acc": 0.9307073030477285
},
{
"epoch": 0.10864978902953587,
"grad_norm": 0.87109375,
"learning_rate": 9.969832602462174e-07,
"loss": 0.22050908207893372,
"step": 515,
"token_acc": 0.9343434343434344
},
{
"epoch": 0.10886075949367088,
"grad_norm": 0.7421875,
"learning_rate": 9.969646859678256e-07,
"loss": 0.25485992431640625,
"step": 516,
"token_acc": 0.9255828808687321
},
{
"epoch": 0.10907172995780591,
"grad_norm": 0.671875,
"learning_rate": 9.969460548573156e-07,
"loss": 0.24492983520030975,
"step": 517,
"token_acc": 0.9348314606741573
},
{
"epoch": 0.10928270042194092,
"grad_norm": 0.7265625,
"learning_rate": 9.96927366916818e-07,
"loss": 0.28373780846595764,
"step": 518,
"token_acc": 0.9258015267175572
},
{
"epoch": 0.10949367088607595,
"grad_norm": 0.75,
"learning_rate": 9.969086221484701e-07,
"loss": 0.2899026870727539,
"step": 519,
"token_acc": 0.9206824304100568
},
{
"epoch": 0.10970464135021098,
"grad_norm": 0.80078125,
"learning_rate": 9.968898205544153e-07,
"loss": 0.2812725305557251,
"step": 520,
"token_acc": 0.9226502311248074
},
{
"epoch": 0.10991561181434599,
"grad_norm": 0.6171875,
"learning_rate": 9.968709621368041e-07,
"loss": 0.24981510639190674,
"step": 521,
"token_acc": 0.9326704545454545
},
{
"epoch": 0.11012658227848102,
"grad_norm": 0.79296875,
"learning_rate": 9.96852046897793e-07,
"loss": 0.3039143681526184,
"step": 522,
"token_acc": 0.9203347799132052
},
{
"epoch": 0.11033755274261603,
"grad_norm": 0.92578125,
"learning_rate": 9.968330748395448e-07,
"loss": 0.2633418142795563,
"step": 523,
"token_acc": 0.9283835135925168
},
{
"epoch": 0.11054852320675106,
"grad_norm": 1.4609375,
"learning_rate": 9.968140459642294e-07,
"loss": 0.24586576223373413,
"step": 524,
"token_acc": 0.9281601316150261
},
{
"epoch": 0.11075949367088607,
"grad_norm": 0.8515625,
"learning_rate": 9.967949602740228e-07,
"loss": 0.2739730477333069,
"step": 525,
"token_acc": 0.9166417687481326
},
{
"epoch": 0.1109704641350211,
"grad_norm": 0.796875,
"learning_rate": 9.967758177711076e-07,
"loss": 0.2627703845500946,
"step": 526,
"token_acc": 0.9227409227409228
},
{
"epoch": 0.11118143459915612,
"grad_norm": 0.80859375,
"learning_rate": 9.967566184576732e-07,
"loss": 0.26023009419441223,
"step": 527,
"token_acc": 0.927381745502998
},
{
"epoch": 0.11139240506329114,
"grad_norm": 0.78125,
"learning_rate": 9.967373623359148e-07,
"loss": 0.24462240934371948,
"step": 528,
"token_acc": 0.9283416203568294
},
{
"epoch": 0.11160337552742616,
"grad_norm": 0.72265625,
"learning_rate": 9.967180494080347e-07,
"loss": 0.24981698393821716,
"step": 529,
"token_acc": 0.9291265153870065
},
{
"epoch": 0.11181434599156118,
"grad_norm": 0.78125,
"learning_rate": 9.966986796762414e-07,
"loss": 0.2446298450231552,
"step": 530,
"token_acc": 0.9370728929384966
},
{
"epoch": 0.1120253164556962,
"grad_norm": 2.421875,
"learning_rate": 9.9667925314275e-07,
"loss": 0.24656617641448975,
"step": 531,
"token_acc": 0.9356594110115237
},
{
"epoch": 0.11223628691983123,
"grad_norm": 0.8203125,
"learning_rate": 9.966597698097823e-07,
"loss": 0.2559359073638916,
"step": 532,
"token_acc": 0.9327158812312721
},
{
"epoch": 0.11244725738396624,
"grad_norm": 0.6640625,
"learning_rate": 9.966402296795661e-07,
"loss": 0.2284064143896103,
"step": 533,
"token_acc": 0.9354838709677419
},
{
"epoch": 0.11265822784810127,
"grad_norm": 0.77734375,
"learning_rate": 9.966206327543362e-07,
"loss": 0.2628895938396454,
"step": 534,
"token_acc": 0.923578751164958
},
{
"epoch": 0.11286919831223628,
"grad_norm": 0.6796875,
"learning_rate": 9.966009790363337e-07,
"loss": 0.2363075464963913,
"step": 535,
"token_acc": 0.9275167785234899
},
{
"epoch": 0.11308016877637131,
"grad_norm": 0.90625,
"learning_rate": 9.965812685278059e-07,
"loss": 0.2766547203063965,
"step": 536,
"token_acc": 0.9212160836874795
},
{
"epoch": 0.11329113924050632,
"grad_norm": 0.6875,
"learning_rate": 9.96561501231007e-07,
"loss": 0.24981704354286194,
"step": 537,
"token_acc": 0.9287794545935928
},
{
"epoch": 0.11350210970464135,
"grad_norm": 0.72265625,
"learning_rate": 9.965416771481975e-07,
"loss": 0.2477213591337204,
"step": 538,
"token_acc": 0.9247878255779924
},
{
"epoch": 0.11371308016877638,
"grad_norm": 0.97265625,
"learning_rate": 9.965217962816446e-07,
"loss": 0.2585391104221344,
"step": 539,
"token_acc": 0.9276463963963963
},
{
"epoch": 0.11392405063291139,
"grad_norm": 0.703125,
"learning_rate": 9.965018586336218e-07,
"loss": 0.24559935927391052,
"step": 540,
"token_acc": 0.9349939246658566
},
{
"epoch": 0.11413502109704642,
"grad_norm": 0.72265625,
"learning_rate": 9.96481864206409e-07,
"loss": 0.22781570255756378,
"step": 541,
"token_acc": 0.9306022623051055
},
{
"epoch": 0.11434599156118143,
"grad_norm": 1.0390625,
"learning_rate": 9.964618130022931e-07,
"loss": 0.2166275829076767,
"step": 542,
"token_acc": 0.9374454466104161
},
{
"epoch": 0.11455696202531646,
"grad_norm": 0.703125,
"learning_rate": 9.964417050235665e-07,
"loss": 0.267704039812088,
"step": 543,
"token_acc": 0.9261410788381743
},
{
"epoch": 0.11476793248945148,
"grad_norm": 0.8203125,
"learning_rate": 9.964215402725294e-07,
"loss": 0.23303918540477753,
"step": 544,
"token_acc": 0.9341683658607631
},
{
"epoch": 0.1149789029535865,
"grad_norm": 0.921875,
"learning_rate": 9.964013187514872e-07,
"loss": 0.33097875118255615,
"step": 545,
"token_acc": 0.9122987324426174
},
{
"epoch": 0.11518987341772152,
"grad_norm": 1.9453125,
"learning_rate": 9.963810404627529e-07,
"loss": 0.2524172067642212,
"step": 546,
"token_acc": 0.9373441396508728
},
{
"epoch": 0.11540084388185654,
"grad_norm": 0.60546875,
"learning_rate": 9.963607054086453e-07,
"loss": 0.25729498267173767,
"step": 547,
"token_acc": 0.9215870040612308
},
{
"epoch": 0.11561181434599156,
"grad_norm": 0.75,
"learning_rate": 9.963403135914898e-07,
"loss": 0.2928774356842041,
"step": 548,
"token_acc": 0.9192731605600238
},
{
"epoch": 0.11582278481012659,
"grad_norm": 0.6953125,
"learning_rate": 9.963198650136184e-07,
"loss": 0.25337544083595276,
"step": 549,
"token_acc": 0.9240650870682272
},
{
"epoch": 0.1160337552742616,
"grad_norm": 0.7109375,
"learning_rate": 9.962993596773697e-07,
"loss": 0.27310362458229065,
"step": 550,
"token_acc": 0.9247853124074622
},
{
"epoch": 0.11624472573839663,
"grad_norm": 0.640625,
"learning_rate": 9.962787975850886e-07,
"loss": 0.22571566700935364,
"step": 551,
"token_acc": 0.9384902143522833
},
{
"epoch": 0.11645569620253164,
"grad_norm": 0.8671875,
"learning_rate": 9.962581787391265e-07,
"loss": 0.25049251317977905,
"step": 552,
"token_acc": 0.9287239722370528
},
{
"epoch": 0.11666666666666667,
"grad_norm": 1.2734375,
"learning_rate": 9.962375031418413e-07,
"loss": 0.24676430225372314,
"step": 553,
"token_acc": 0.9325946445060018
},
{
"epoch": 0.11687763713080168,
"grad_norm": 0.75,
"learning_rate": 9.962167707955977e-07,
"loss": 0.22018642723560333,
"step": 554,
"token_acc": 0.9440233236151604
},
{
"epoch": 0.11708860759493671,
"grad_norm": 0.7421875,
"learning_rate": 9.96195981702766e-07,
"loss": 0.2333768904209137,
"step": 555,
"token_acc": 0.9352249928346231
},
{
"epoch": 0.11729957805907174,
"grad_norm": 0.703125,
"learning_rate": 9.961751358657244e-07,
"loss": 0.2830660939216614,
"step": 556,
"token_acc": 0.9188869153345175
},
{
"epoch": 0.11751054852320675,
"grad_norm": 0.81640625,
"learning_rate": 9.961542332868564e-07,
"loss": 0.26290833950042725,
"step": 557,
"token_acc": 0.9261704681872749
},
{
"epoch": 0.11772151898734177,
"grad_norm": 0.7421875,
"learning_rate": 9.961332739685523e-07,
"loss": 0.2768633961677551,
"step": 558,
"token_acc": 0.9245508982035928
},
{
"epoch": 0.11793248945147679,
"grad_norm": 0.80078125,
"learning_rate": 9.96112257913209e-07,
"loss": 0.2084287852048874,
"step": 559,
"token_acc": 0.9381261048909841
},
{
"epoch": 0.11814345991561181,
"grad_norm": 0.796875,
"learning_rate": 9.960911851232301e-07,
"loss": 0.2791953682899475,
"step": 560,
"token_acc": 0.924936025021325
},
{
"epoch": 0.11835443037974684,
"grad_norm": 1.34375,
"learning_rate": 9.960700556010253e-07,
"loss": 0.319602370262146,
"step": 561,
"token_acc": 0.9191286183228887
},
{
"epoch": 0.11856540084388185,
"grad_norm": 0.8515625,
"learning_rate": 9.960488693490108e-07,
"loss": 0.21053284406661987,
"step": 562,
"token_acc": 0.9407194244604317
},
{
"epoch": 0.11877637130801688,
"grad_norm": 0.7421875,
"learning_rate": 9.960276263696097e-07,
"loss": 0.27438345551490784,
"step": 563,
"token_acc": 0.9290875033449291
},
{
"epoch": 0.1189873417721519,
"grad_norm": 0.6328125,
"learning_rate": 9.960063266652512e-07,
"loss": 0.2969055771827698,
"step": 564,
"token_acc": 0.918200408997955
},
{
"epoch": 0.11919831223628692,
"grad_norm": 1.0234375,
"learning_rate": 9.95984970238371e-07,
"loss": 0.3027820587158203,
"step": 565,
"token_acc": 0.924516531503431
},
{
"epoch": 0.11940928270042193,
"grad_norm": 0.69921875,
"learning_rate": 9.959635570914115e-07,
"loss": 0.26206478476524353,
"step": 566,
"token_acc": 0.9267332727823191
},
{
"epoch": 0.11962025316455696,
"grad_norm": 0.8359375,
"learning_rate": 9.959420872268214e-07,
"loss": 0.22268003225326538,
"step": 567,
"token_acc": 0.9377475947934352
},
{
"epoch": 0.11983122362869199,
"grad_norm": 0.72265625,
"learning_rate": 9.95920560647056e-07,
"loss": 0.2821509838104248,
"step": 568,
"token_acc": 0.9257203277821835
},
{
"epoch": 0.120042194092827,
"grad_norm": 0.8125,
"learning_rate": 9.958989773545772e-07,
"loss": 0.2774399518966675,
"step": 569,
"token_acc": 0.9208851167020309
},
{
"epoch": 0.12025316455696203,
"grad_norm": 0.73046875,
"learning_rate": 9.95877337351853e-07,
"loss": 0.20950725674629211,
"step": 570,
"token_acc": 0.9383966244725739
},
{
"epoch": 0.12046413502109704,
"grad_norm": 2.375,
"learning_rate": 9.95855640641358e-07,
"loss": 0.24889153242111206,
"step": 571,
"token_acc": 0.9338555265448216
},
{
"epoch": 0.12067510548523207,
"grad_norm": 0.7265625,
"learning_rate": 9.958338872255738e-07,
"loss": 0.27347537875175476,
"step": 572,
"token_acc": 0.9261669024045261
},
{
"epoch": 0.1208860759493671,
"grad_norm": 0.625,
"learning_rate": 9.958120771069878e-07,
"loss": 0.2640995383262634,
"step": 573,
"token_acc": 0.9278178789561354
},
{
"epoch": 0.12109704641350211,
"grad_norm": 0.609375,
"learning_rate": 9.957902102880945e-07,
"loss": 0.23652713000774384,
"step": 574,
"token_acc": 0.9364988558352403
},
{
"epoch": 0.12130801687763713,
"grad_norm": 0.8359375,
"learning_rate": 9.957682867713942e-07,
"loss": 0.291990727186203,
"step": 575,
"token_acc": 0.9223826714801444
},
{
"epoch": 0.12151898734177215,
"grad_norm": 0.6328125,
"learning_rate": 9.95746306559394e-07,
"loss": 0.21727707982063293,
"step": 576,
"token_acc": 0.9308996088657105
},
{
"epoch": 0.12172995780590717,
"grad_norm": 0.66796875,
"learning_rate": 9.957242696546077e-07,
"loss": 0.2906607985496521,
"step": 577,
"token_acc": 0.9158725837190308
},
{
"epoch": 0.1219409282700422,
"grad_norm": 0.7265625,
"learning_rate": 9.957021760595556e-07,
"loss": 0.226593479514122,
"step": 578,
"token_acc": 0.9271303824149353
},
{
"epoch": 0.12215189873417721,
"grad_norm": 0.7265625,
"learning_rate": 9.956800257767639e-07,
"loss": 0.26656001806259155,
"step": 579,
"token_acc": 0.930952380952381
},
{
"epoch": 0.12236286919831224,
"grad_norm": 0.6875,
"learning_rate": 9.956578188087658e-07,
"loss": 0.2880259156227112,
"step": 580,
"token_acc": 0.9256572982774252
},
{
"epoch": 0.12257383966244725,
"grad_norm": 0.83984375,
"learning_rate": 9.95635555158101e-07,
"loss": 0.24349641799926758,
"step": 581,
"token_acc": 0.9316569954867827
},
{
"epoch": 0.12278481012658228,
"grad_norm": 0.7109375,
"learning_rate": 9.956132348273157e-07,
"loss": 0.24496349692344666,
"step": 582,
"token_acc": 0.9309711286089238
},
{
"epoch": 0.1229957805907173,
"grad_norm": 0.75390625,
"learning_rate": 9.955908578189619e-07,
"loss": 0.2652456760406494,
"step": 583,
"token_acc": 0.9269628727936701
},
{
"epoch": 0.12320675105485232,
"grad_norm": 0.87109375,
"learning_rate": 9.955684241355988e-07,
"loss": 0.2777090072631836,
"step": 584,
"token_acc": 0.9167887489012599
},
{
"epoch": 0.12341772151898735,
"grad_norm": 1.2109375,
"learning_rate": 9.95545933779792e-07,
"loss": 0.3164791464805603,
"step": 585,
"token_acc": 0.919442072302875
},
{
"epoch": 0.12362869198312236,
"grad_norm": 0.67578125,
"learning_rate": 9.955233867541134e-07,
"loss": 0.26809951663017273,
"step": 586,
"token_acc": 0.9227027027027027
},
{
"epoch": 0.12383966244725739,
"grad_norm": 0.70703125,
"learning_rate": 9.955007830611414e-07,
"loss": 0.25988298654556274,
"step": 587,
"token_acc": 0.9277679100059206
},
{
"epoch": 0.1240506329113924,
"grad_norm": 1.453125,
"learning_rate": 9.954781227034612e-07,
"loss": 0.22518092393875122,
"step": 588,
"token_acc": 0.935474701534963
},
{
"epoch": 0.12426160337552743,
"grad_norm": 0.62890625,
"learning_rate": 9.954554056836637e-07,
"loss": 0.23757173120975494,
"step": 589,
"token_acc": 0.929593589009731
},
{
"epoch": 0.12447257383966245,
"grad_norm": 0.7734375,
"learning_rate": 9.954326320043472e-07,
"loss": 0.2483949363231659,
"step": 590,
"token_acc": 0.9315320847405588
},
{
"epoch": 0.12468354430379747,
"grad_norm": 0.82421875,
"learning_rate": 9.95409801668116e-07,
"loss": 0.26530349254608154,
"step": 591,
"token_acc": 0.9214407260351674
},
{
"epoch": 0.1248945147679325,
"grad_norm": 0.703125,
"learning_rate": 9.953869146775806e-07,
"loss": 0.2826001048088074,
"step": 592,
"token_acc": 0.9162200282087447
},
{
"epoch": 0.12510548523206752,
"grad_norm": 0.59375,
"learning_rate": 9.953639710353589e-07,
"loss": 0.1961941421031952,
"step": 593,
"token_acc": 0.9426091825307951
},
{
"epoch": 0.12531645569620253,
"grad_norm": 0.82421875,
"learning_rate": 9.953409707440742e-07,
"loss": 0.26104363799095154,
"step": 594,
"token_acc": 0.9215813350615684
},
{
"epoch": 0.12552742616033755,
"grad_norm": 0.828125,
"learning_rate": 9.95317913806357e-07,
"loss": 0.30334994196891785,
"step": 595,
"token_acc": 0.9229497354497355
},
{
"epoch": 0.1257383966244726,
"grad_norm": 0.8125,
"learning_rate": 9.95294800224844e-07,
"loss": 0.27462461590766907,
"step": 596,
"token_acc": 0.9222963177732676
},
{
"epoch": 0.1259493670886076,
"grad_norm": 0.68359375,
"learning_rate": 9.952716300021784e-07,
"loss": 0.25919121503829956,
"step": 597,
"token_acc": 0.9274289099526066
},
{
"epoch": 0.1261603375527426,
"grad_norm": 0.6328125,
"learning_rate": 9.952484031410102e-07,
"loss": 0.24202126264572144,
"step": 598,
"token_acc": 0.9379543094496365
},
{
"epoch": 0.12637130801687763,
"grad_norm": 0.703125,
"learning_rate": 9.95225119643995e-07,
"loss": 0.22879423201084137,
"step": 599,
"token_acc": 0.9303405572755418
},
{
"epoch": 0.12658227848101267,
"grad_norm": 0.66015625,
"learning_rate": 9.952017795137962e-07,
"loss": 0.2557697892189026,
"step": 600,
"token_acc": 0.9307598039215687
},
{
"epoch": 0.12658227848101267,
"eval_loss": 0.4336538016796112,
"eval_runtime": 246.0329,
"eval_samples_per_second": 136.994,
"eval_steps_per_second": 2.142,
"eval_token_acc": 0.899098088486045,
"step": 600
},
{
"epoch": 0.12679324894514768,
"grad_norm": 0.62109375,
"learning_rate": 9.951783827530821e-07,
"loss": 0.24460609257221222,
"step": 601,
"token_acc": 0.935226264418811
},
{
"epoch": 0.1270042194092827,
"grad_norm": 0.65625,
"learning_rate": 9.951549293645292e-07,
"loss": 0.24832656979560852,
"step": 602,
"token_acc": 0.9309408926417371
},
{
"epoch": 0.12721518987341773,
"grad_norm": 0.69140625,
"learning_rate": 9.95131419350819e-07,
"loss": 0.23030316829681396,
"step": 603,
"token_acc": 0.9332206255283179
},
{
"epoch": 0.12742616033755275,
"grad_norm": 0.73046875,
"learning_rate": 9.951078527146403e-07,
"loss": 0.2880566418170929,
"step": 604,
"token_acc": 0.9192907367777438
},
{
"epoch": 0.12763713080168776,
"grad_norm": 0.6796875,
"learning_rate": 9.95084229458688e-07,
"loss": 0.24888081848621368,
"step": 605,
"token_acc": 0.929957805907173
},
{
"epoch": 0.12784810126582277,
"grad_norm": 0.609375,
"learning_rate": 9.950605495856637e-07,
"loss": 0.2833850681781769,
"step": 606,
"token_acc": 0.9213352685050799
},
{
"epoch": 0.1280590717299578,
"grad_norm": 0.875,
"learning_rate": 9.950368130982755e-07,
"loss": 0.26693737506866455,
"step": 607,
"token_acc": 0.9250070482097548
},
{
"epoch": 0.12827004219409283,
"grad_norm": 0.62890625,
"learning_rate": 9.950130199992377e-07,
"loss": 0.23543164134025574,
"step": 608,
"token_acc": 0.9348515422311905
},
{
"epoch": 0.12848101265822784,
"grad_norm": 0.7265625,
"learning_rate": 9.949891702912712e-07,
"loss": 0.22989103198051453,
"step": 609,
"token_acc": 0.9318626082099972
},
{
"epoch": 0.12869198312236288,
"grad_norm": 0.74609375,
"learning_rate": 9.949652639771036e-07,
"loss": 0.24984115362167358,
"step": 610,
"token_acc": 0.9261559696342305
},
{
"epoch": 0.1289029535864979,
"grad_norm": 0.70703125,
"learning_rate": 9.94941301059469e-07,
"loss": 0.2549693286418915,
"step": 611,
"token_acc": 0.9259363559560687
},
{
"epoch": 0.1291139240506329,
"grad_norm": 0.66015625,
"learning_rate": 9.94917281541107e-07,
"loss": 0.2592216432094574,
"step": 612,
"token_acc": 0.9255747126436782
},
{
"epoch": 0.12932489451476795,
"grad_norm": 0.8046875,
"learning_rate": 9.948932054247652e-07,
"loss": 0.2784273624420166,
"step": 613,
"token_acc": 0.9198324022346369
},
{
"epoch": 0.12953586497890296,
"grad_norm": 1.359375,
"learning_rate": 9.948690727131965e-07,
"loss": 0.2754824161529541,
"step": 614,
"token_acc": 0.9211413748378728
},
{
"epoch": 0.12974683544303797,
"grad_norm": 0.62890625,
"learning_rate": 9.948448834091608e-07,
"loss": 0.22421778738498688,
"step": 615,
"token_acc": 0.9337539432176656
},
{
"epoch": 0.12995780590717299,
"grad_norm": 0.6640625,
"learning_rate": 9.948206375154244e-07,
"loss": 0.22916918992996216,
"step": 616,
"token_acc": 0.933944374209861
},
{
"epoch": 0.13016877637130803,
"grad_norm": 0.640625,
"learning_rate": 9.947963350347598e-07,
"loss": 0.23158694803714752,
"step": 617,
"token_acc": 0.9291338582677166
},
{
"epoch": 0.13037974683544304,
"grad_norm": 0.85546875,
"learning_rate": 9.947719759699466e-07,
"loss": 0.2788570523262024,
"step": 618,
"token_acc": 0.9231003039513678
},
{
"epoch": 0.13059071729957805,
"grad_norm": 0.69140625,
"learning_rate": 9.947475603237702e-07,
"loss": 0.28004133701324463,
"step": 619,
"token_acc": 0.9180544541369283
},
{
"epoch": 0.1308016877637131,
"grad_norm": 0.71484375,
"learning_rate": 9.947230880990227e-07,
"loss": 0.2773160934448242,
"step": 620,
"token_acc": 0.9248013620885358
},
{
"epoch": 0.1310126582278481,
"grad_norm": 0.71484375,
"learning_rate": 9.946985592985028e-07,
"loss": 0.2508021593093872,
"step": 621,
"token_acc": 0.9334818586887333
},
{
"epoch": 0.13122362869198312,
"grad_norm": 0.8046875,
"learning_rate": 9.946739739250156e-07,
"loss": 0.23769596219062805,
"step": 622,
"token_acc": 0.9302249755461363
},
{
"epoch": 0.13143459915611813,
"grad_norm": 1.2109375,
"learning_rate": 9.946493319813725e-07,
"loss": 0.21937592327594757,
"step": 623,
"token_acc": 0.9373860182370821
},
{
"epoch": 0.13164556962025317,
"grad_norm": 0.62109375,
"learning_rate": 9.946246334703916e-07,
"loss": 0.27754154801368713,
"step": 624,
"token_acc": 0.9261245159368484
},
{
"epoch": 0.13185654008438819,
"grad_norm": 0.78515625,
"learning_rate": 9.945998783948975e-07,
"loss": 0.2924942672252655,
"step": 625,
"token_acc": 0.9139688249400479
},
{
"epoch": 0.1320675105485232,
"grad_norm": 0.64453125,
"learning_rate": 9.945750667577209e-07,
"loss": 0.2303755283355713,
"step": 626,
"token_acc": 0.9341597796143251
},
{
"epoch": 0.13227848101265824,
"grad_norm": 0.7734375,
"learning_rate": 9.945501985616995e-07,
"loss": 0.20424559712409973,
"step": 627,
"token_acc": 0.9430594900849858
},
{
"epoch": 0.13248945147679325,
"grad_norm": 0.64453125,
"learning_rate": 9.94525273809677e-07,
"loss": 0.2620590031147003,
"step": 628,
"token_acc": 0.9274099883855982
},
{
"epoch": 0.13270042194092826,
"grad_norm": 0.76171875,
"learning_rate": 9.945002925045038e-07,
"loss": 0.2684752643108368,
"step": 629,
"token_acc": 0.9270194986072423
},
{
"epoch": 0.13291139240506328,
"grad_norm": 0.70703125,
"learning_rate": 9.944752546490367e-07,
"loss": 0.23374760150909424,
"step": 630,
"token_acc": 0.9325842696629213
},
{
"epoch": 0.13312236286919832,
"grad_norm": 1.1953125,
"learning_rate": 9.94450160246139e-07,
"loss": 0.22228139638900757,
"step": 631,
"token_acc": 0.9354534005037783
},
{
"epoch": 0.13333333333333333,
"grad_norm": 0.73046875,
"learning_rate": 9.944250092986807e-07,
"loss": 0.2018851488828659,
"step": 632,
"token_acc": 0.9405840886203424
},
{
"epoch": 0.13354430379746834,
"grad_norm": 1.1796875,
"learning_rate": 9.943998018095377e-07,
"loss": 0.26342812180519104,
"step": 633,
"token_acc": 0.9279495646952867
},
{
"epoch": 0.13375527426160339,
"grad_norm": 0.734375,
"learning_rate": 9.943745377815927e-07,
"loss": 0.24731256067752838,
"step": 634,
"token_acc": 0.9329608938547486
},
{
"epoch": 0.1339662447257384,
"grad_norm": 0.72265625,
"learning_rate": 9.94349217217735e-07,
"loss": 0.22763285040855408,
"step": 635,
"token_acc": 0.9298531810766721
},
{
"epoch": 0.1341772151898734,
"grad_norm": 0.72265625,
"learning_rate": 9.943238401208602e-07,
"loss": 0.25396978855133057,
"step": 636,
"token_acc": 0.9292196007259528
},
{
"epoch": 0.13438818565400845,
"grad_norm": 0.7421875,
"learning_rate": 9.942984064938705e-07,
"loss": 0.30096304416656494,
"step": 637,
"token_acc": 0.9151027703306523
},
{
"epoch": 0.13459915611814346,
"grad_norm": 0.6484375,
"learning_rate": 9.942729163396741e-07,
"loss": 0.29584643244743347,
"step": 638,
"token_acc": 0.919442072302875
},
{
"epoch": 0.13481012658227848,
"grad_norm": 0.65625,
"learning_rate": 9.942473696611862e-07,
"loss": 0.2260875701904297,
"step": 639,
"token_acc": 0.934560327198364
},
{
"epoch": 0.1350210970464135,
"grad_norm": 0.703125,
"learning_rate": 9.942217664613284e-07,
"loss": 0.25592464208602905,
"step": 640,
"token_acc": 0.926865671641791
},
{
"epoch": 0.13523206751054853,
"grad_norm": 0.6171875,
"learning_rate": 9.941961067430285e-07,
"loss": 0.24965469539165497,
"step": 641,
"token_acc": 0.92850705917693
},
{
"epoch": 0.13544303797468354,
"grad_norm": 0.7109375,
"learning_rate": 9.94170390509221e-07,
"loss": 0.2692350149154663,
"step": 642,
"token_acc": 0.9312614259597807
},
{
"epoch": 0.13565400843881856,
"grad_norm": 0.78125,
"learning_rate": 9.941446177628467e-07,
"loss": 0.2497376799583435,
"step": 643,
"token_acc": 0.925979262672811
},
{
"epoch": 0.1358649789029536,
"grad_norm": 1.046875,
"learning_rate": 9.94118788506853e-07,
"loss": 0.27541112899780273,
"step": 644,
"token_acc": 0.9211567732115677
},
{
"epoch": 0.1360759493670886,
"grad_norm": 0.6953125,
"learning_rate": 9.940929027441936e-07,
"loss": 0.2414344996213913,
"step": 645,
"token_acc": 0.9313432835820895
},
{
"epoch": 0.13628691983122362,
"grad_norm": 0.62109375,
"learning_rate": 9.940669604778288e-07,
"loss": 0.2977514863014221,
"step": 646,
"token_acc": 0.922911547911548
},
{
"epoch": 0.13649789029535864,
"grad_norm": 0.71484375,
"learning_rate": 9.940409617107252e-07,
"loss": 0.2718917727470398,
"step": 647,
"token_acc": 0.9186241610738255
},
{
"epoch": 0.13670886075949368,
"grad_norm": 0.765625,
"learning_rate": 9.940149064458563e-07,
"loss": 0.2234637290239334,
"step": 648,
"token_acc": 0.9408866995073891
},
{
"epoch": 0.1369198312236287,
"grad_norm": 0.640625,
"learning_rate": 9.939887946862017e-07,
"loss": 0.2774735689163208,
"step": 649,
"token_acc": 0.9290869155946031
},
{
"epoch": 0.1371308016877637,
"grad_norm": 0.90625,
"learning_rate": 9.93962626434747e-07,
"loss": 0.2615464925765991,
"step": 650,
"token_acc": 0.9272910881090634
},
{
"epoch": 0.13734177215189874,
"grad_norm": 0.65625,
"learning_rate": 9.939364016944852e-07,
"loss": 0.23573726415634155,
"step": 651,
"token_acc": 0.9329399141630901
},
{
"epoch": 0.13755274261603376,
"grad_norm": 0.6328125,
"learning_rate": 9.939101204684151e-07,
"loss": 0.2487039864063263,
"step": 652,
"token_acc": 0.932972972972973
},
{
"epoch": 0.13776371308016877,
"grad_norm": 0.6875,
"learning_rate": 9.938837827595424e-07,
"loss": 0.26214438676834106,
"step": 653,
"token_acc": 0.9278832116788321
},
{
"epoch": 0.1379746835443038,
"grad_norm": 0.69921875,
"learning_rate": 9.938573885708792e-07,
"loss": 0.24997225403785706,
"step": 654,
"token_acc": 0.9322553666016169
},
{
"epoch": 0.13818565400843882,
"grad_norm": 0.76953125,
"learning_rate": 9.938309379054433e-07,
"loss": 0.28863316774368286,
"step": 655,
"token_acc": 0.9217944831767202
},
{
"epoch": 0.13839662447257384,
"grad_norm": 0.625,
"learning_rate": 9.9380443076626e-07,
"loss": 0.23938237130641937,
"step": 656,
"token_acc": 0.9307647740440325
},
{
"epoch": 0.13860759493670885,
"grad_norm": 0.703125,
"learning_rate": 9.937778671563606e-07,
"loss": 0.26946017146110535,
"step": 657,
"token_acc": 0.9250278706800446
},
{
"epoch": 0.1388185654008439,
"grad_norm": 0.64453125,
"learning_rate": 9.937512470787827e-07,
"loss": 0.25879329442977905,
"step": 658,
"token_acc": 0.9263622974963182
},
{
"epoch": 0.1390295358649789,
"grad_norm": 0.75390625,
"learning_rate": 9.937245705365707e-07,
"loss": 0.26367712020874023,
"step": 659,
"token_acc": 0.9273255813953488
},
{
"epoch": 0.13924050632911392,
"grad_norm": 0.73828125,
"learning_rate": 9.93697837532775e-07,
"loss": 0.27003130316734314,
"step": 660,
"token_acc": 0.9255381035485748
},
{
"epoch": 0.13945147679324896,
"grad_norm": 0.77734375,
"learning_rate": 9.936710480704531e-07,
"loss": 0.3241864740848541,
"step": 661,
"token_acc": 0.9117466174661747
},
{
"epoch": 0.13966244725738397,
"grad_norm": 0.765625,
"learning_rate": 9.936442021526685e-07,
"loss": 0.254525363445282,
"step": 662,
"token_acc": 0.9274787535410765
},
{
"epoch": 0.13987341772151898,
"grad_norm": 0.85546875,
"learning_rate": 9.936172997824912e-07,
"loss": 0.22039127349853516,
"step": 663,
"token_acc": 0.9337885985748219
},
{
"epoch": 0.140084388185654,
"grad_norm": 0.71484375,
"learning_rate": 9.935903409629977e-07,
"loss": 0.26330018043518066,
"step": 664,
"token_acc": 0.9245566576495341
},
{
"epoch": 0.14029535864978904,
"grad_norm": 0.79296875,
"learning_rate": 9.93563325697271e-07,
"loss": 0.25053250789642334,
"step": 665,
"token_acc": 0.9321644150617994
},
{
"epoch": 0.14050632911392405,
"grad_norm": 1.015625,
"learning_rate": 9.935362539884004e-07,
"loss": 0.23359492421150208,
"step": 666,
"token_acc": 0.9295649600473513
},
{
"epoch": 0.14071729957805906,
"grad_norm": 0.80078125,
"learning_rate": 9.935091258394821e-07,
"loss": 0.3050011098384857,
"step": 667,
"token_acc": 0.9222160044767768
},
{
"epoch": 0.1409282700421941,
"grad_norm": 0.65625,
"learning_rate": 9.93481941253618e-07,
"loss": 0.24552345275878906,
"step": 668,
"token_acc": 0.9327153110047847
},
{
"epoch": 0.14113924050632912,
"grad_norm": 0.6328125,
"learning_rate": 9.934547002339174e-07,
"loss": 0.2593832015991211,
"step": 669,
"token_acc": 0.9236835410836938
},
{
"epoch": 0.14135021097046413,
"grad_norm": 0.765625,
"learning_rate": 9.93427402783495e-07,
"loss": 0.2546125054359436,
"step": 670,
"token_acc": 0.9320175438596491
},
{
"epoch": 0.14156118143459914,
"grad_norm": 0.84765625,
"learning_rate": 9.93400048905473e-07,
"loss": 0.27444595098495483,
"step": 671,
"token_acc": 0.9244940321743643
},
{
"epoch": 0.14177215189873418,
"grad_norm": 0.79296875,
"learning_rate": 9.93372638602979e-07,
"loss": 0.2675279378890991,
"step": 672,
"token_acc": 0.9254424136930665
},
{
"epoch": 0.1419831223628692,
"grad_norm": 0.63671875,
"learning_rate": 9.933451718791481e-07,
"loss": 0.22922030091285706,
"step": 673,
"token_acc": 0.9329147389292796
},
{
"epoch": 0.1421940928270042,
"grad_norm": 1.015625,
"learning_rate": 9.933176487371213e-07,
"loss": 0.3030126094818115,
"step": 674,
"token_acc": 0.9166399487015069
},
{
"epoch": 0.14240506329113925,
"grad_norm": 0.76171875,
"learning_rate": 9.932900691800457e-07,
"loss": 0.2756281793117523,
"step": 675,
"token_acc": 0.921146953405018
},
{
"epoch": 0.14261603375527426,
"grad_norm": 0.6875,
"learning_rate": 9.932624332110758e-07,
"loss": 0.23098278045654297,
"step": 676,
"token_acc": 0.937677859988617
},
{
"epoch": 0.14282700421940928,
"grad_norm": 0.58203125,
"learning_rate": 9.932347408333715e-07,
"loss": 0.22887524962425232,
"step": 677,
"token_acc": 0.9341611319665031
},
{
"epoch": 0.14303797468354432,
"grad_norm": 0.98046875,
"learning_rate": 9.932069920501e-07,
"loss": 0.2759955823421478,
"step": 678,
"token_acc": 0.9240579710144927
},
{
"epoch": 0.14324894514767933,
"grad_norm": 0.59375,
"learning_rate": 9.931791868644341e-07,
"loss": 0.2028590440750122,
"step": 679,
"token_acc": 0.9378352792679079
},
{
"epoch": 0.14345991561181434,
"grad_norm": 0.8203125,
"learning_rate": 9.931513252795543e-07,
"loss": 0.30346542596817017,
"step": 680,
"token_acc": 0.9129104062326099
},
{
"epoch": 0.14367088607594936,
"grad_norm": 0.98828125,
"learning_rate": 9.931234072986466e-07,
"loss": 0.27435851097106934,
"step": 681,
"token_acc": 0.9276353276353276
},
{
"epoch": 0.1438818565400844,
"grad_norm": 0.83203125,
"learning_rate": 9.930954329249032e-07,
"loss": 0.2799455523490906,
"step": 682,
"token_acc": 0.9241399588356366
},
{
"epoch": 0.1440928270042194,
"grad_norm": 0.82421875,
"learning_rate": 9.930674021615237e-07,
"loss": 0.28436923027038574,
"step": 683,
"token_acc": 0.9250146455770357
},
{
"epoch": 0.14430379746835442,
"grad_norm": 0.69921875,
"learning_rate": 9.930393150117133e-07,
"loss": 0.29506832361221313,
"step": 684,
"token_acc": 0.9232
},
{
"epoch": 0.14451476793248946,
"grad_norm": 0.66796875,
"learning_rate": 9.930111714786844e-07,
"loss": 0.27069365978240967,
"step": 685,
"token_acc": 0.926836079307456
},
{
"epoch": 0.14472573839662448,
"grad_norm": 0.78125,
"learning_rate": 9.92982971565655e-07,
"loss": 0.21447613835334778,
"step": 686,
"token_acc": 0.9363425925925926
},
{
"epoch": 0.1449367088607595,
"grad_norm": 0.66796875,
"learning_rate": 9.929547152758505e-07,
"loss": 0.2686905264854431,
"step": 687,
"token_acc": 0.9303255282695603
},
{
"epoch": 0.1451476793248945,
"grad_norm": 0.71484375,
"learning_rate": 9.929264026125017e-07,
"loss": 0.27938973903656006,
"step": 688,
"token_acc": 0.92536881689326
},
{
"epoch": 0.14535864978902954,
"grad_norm": 0.6953125,
"learning_rate": 9.928980335788469e-07,
"loss": 0.2390938103199005,
"step": 689,
"token_acc": 0.9279638490164805
},
{
"epoch": 0.14556962025316456,
"grad_norm": 0.69921875,
"learning_rate": 9.928696081781299e-07,
"loss": 0.2756063640117645,
"step": 690,
"token_acc": 0.920317553660688
},
{
"epoch": 0.14578059071729957,
"grad_norm": 0.76171875,
"learning_rate": 9.928411264136017e-07,
"loss": 0.23743261396884918,
"step": 691,
"token_acc": 0.9318757921419518
},
{
"epoch": 0.1459915611814346,
"grad_norm": 0.70703125,
"learning_rate": 9.928125882885193e-07,
"loss": 0.2753446102142334,
"step": 692,
"token_acc": 0.9266131265577402
},
{
"epoch": 0.14620253164556962,
"grad_norm": 0.8359375,
"learning_rate": 9.927839938061461e-07,
"loss": 0.24559064209461212,
"step": 693,
"token_acc": 0.9276517473942366
},
{
"epoch": 0.14641350210970464,
"grad_norm": 0.6640625,
"learning_rate": 9.927553429697526e-07,
"loss": 0.24906222522258759,
"step": 694,
"token_acc": 0.9354838709677419
},
{
"epoch": 0.14662447257383968,
"grad_norm": 0.73828125,
"learning_rate": 9.92726635782615e-07,
"loss": 0.26196521520614624,
"step": 695,
"token_acc": 0.9294971487817522
},
{
"epoch": 0.1468354430379747,
"grad_norm": 0.69921875,
"learning_rate": 9.92697872248016e-07,
"loss": 0.28846031427383423,
"step": 696,
"token_acc": 0.9216428779493154
},
{
"epoch": 0.1470464135021097,
"grad_norm": 0.71875,
"learning_rate": 9.926690523692454e-07,
"loss": 0.2781599164009094,
"step": 697,
"token_acc": 0.9191949534394713
},
{
"epoch": 0.14725738396624471,
"grad_norm": 0.6953125,
"learning_rate": 9.926401761495986e-07,
"loss": 0.24464154243469238,
"step": 698,
"token_acc": 0.9295774647887324
},
{
"epoch": 0.14746835443037976,
"grad_norm": 0.6328125,
"learning_rate": 9.926112435923778e-07,
"loss": 0.24627582728862762,
"step": 699,
"token_acc": 0.9308590242442383
},
{
"epoch": 0.14767932489451477,
"grad_norm": 0.80078125,
"learning_rate": 9.92582254700892e-07,
"loss": 0.27795839309692383,
"step": 700,
"token_acc": 0.9238838084991932
},
{
"epoch": 0.14789029535864978,
"grad_norm": 0.62890625,
"learning_rate": 9.925532094784563e-07,
"loss": 0.24588271975517273,
"step": 701,
"token_acc": 0.9284253578732107
},
{
"epoch": 0.14810126582278482,
"grad_norm": 0.66796875,
"learning_rate": 9.92524107928392e-07,
"loss": 0.24630197882652283,
"step": 702,
"token_acc": 0.9327267714364489
},
{
"epoch": 0.14831223628691984,
"grad_norm": 0.7421875,
"learning_rate": 9.924949500540275e-07,
"loss": 0.2578659653663635,
"step": 703,
"token_acc": 0.9267192784667418
},
{
"epoch": 0.14852320675105485,
"grad_norm": 0.671875,
"learning_rate": 9.924657358586967e-07,
"loss": 0.25091686844825745,
"step": 704,
"token_acc": 0.9329545454545455
},
{
"epoch": 0.14873417721518986,
"grad_norm": 0.71484375,
"learning_rate": 9.924364653457411e-07,
"loss": 0.2511135935783386,
"step": 705,
"token_acc": 0.9301768055139347
},
{
"epoch": 0.1489451476793249,
"grad_norm": 0.78125,
"learning_rate": 9.924071385185075e-07,
"loss": 0.2616545259952545,
"step": 706,
"token_acc": 0.927741935483871
},
{
"epoch": 0.14915611814345991,
"grad_norm": 0.68359375,
"learning_rate": 9.9237775538035e-07,
"loss": 0.2902517318725586,
"step": 707,
"token_acc": 0.9226774379688402
},
{
"epoch": 0.14936708860759493,
"grad_norm": 0.7421875,
"learning_rate": 9.92348315934629e-07,
"loss": 0.27046293020248413,
"step": 708,
"token_acc": 0.9296824368114064
},
{
"epoch": 0.14957805907172997,
"grad_norm": 0.62109375,
"learning_rate": 9.923188201847107e-07,
"loss": 0.20588457584381104,
"step": 709,
"token_acc": 0.9350493864112541
},
{
"epoch": 0.14978902953586498,
"grad_norm": 0.83984375,
"learning_rate": 9.92289268133968e-07,
"loss": 0.25359445810317993,
"step": 710,
"token_acc": 0.9285503395335105
},
{
"epoch": 0.15,
"grad_norm": 0.77734375,
"learning_rate": 9.922596597857811e-07,
"loss": 0.267612099647522,
"step": 711,
"token_acc": 0.9265569917743831
},
{
"epoch": 0.150210970464135,
"grad_norm": 0.9140625,
"learning_rate": 9.922299951435357e-07,
"loss": 0.2501794993877411,
"step": 712,
"token_acc": 0.9300184162062615
},
{
"epoch": 0.15042194092827005,
"grad_norm": 0.63671875,
"learning_rate": 9.922002742106242e-07,
"loss": 0.2614431381225586,
"step": 713,
"token_acc": 0.9250471825289835
},
{
"epoch": 0.15063291139240506,
"grad_norm": 0.61328125,
"learning_rate": 9.921704969904453e-07,
"loss": 0.2227068841457367,
"step": 714,
"token_acc": 0.934162192709805
},
{
"epoch": 0.15084388185654007,
"grad_norm": 0.95703125,
"learning_rate": 9.92140663486404e-07,
"loss": 0.2870650887489319,
"step": 715,
"token_acc": 0.9198347107438016
},
{
"epoch": 0.15105485232067511,
"grad_norm": 0.59765625,
"learning_rate": 9.92110773701913e-07,
"loss": 0.24414213001728058,
"step": 716,
"token_acc": 0.9290909090909091
},
{
"epoch": 0.15126582278481013,
"grad_norm": 0.6328125,
"learning_rate": 9.920808276403893e-07,
"loss": 0.27001482248306274,
"step": 717,
"token_acc": 0.9276070094800345
},
{
"epoch": 0.15147679324894514,
"grad_norm": 0.7265625,
"learning_rate": 9.920508253052584e-07,
"loss": 0.24048057198524475,
"step": 718,
"token_acc": 0.9305245535714286
},
{
"epoch": 0.15168776371308018,
"grad_norm": 0.9921875,
"learning_rate": 9.92020766699951e-07,
"loss": 0.267702579498291,
"step": 719,
"token_acc": 0.9254603916983338
},
{
"epoch": 0.1518987341772152,
"grad_norm": 0.66015625,
"learning_rate": 9.919906518279043e-07,
"loss": 0.23744544386863708,
"step": 720,
"token_acc": 0.9312857886517438
},
{
"epoch": 0.1521097046413502,
"grad_norm": 0.85546875,
"learning_rate": 9.919604806925623e-07,
"loss": 0.2514658570289612,
"step": 721,
"token_acc": 0.9293015332197615
},
{
"epoch": 0.15232067510548522,
"grad_norm": 0.7109375,
"learning_rate": 9.919302532973754e-07,
"loss": 0.2536316215991974,
"step": 722,
"token_acc": 0.9287620064034151
},
{
"epoch": 0.15253164556962026,
"grad_norm": 0.8359375,
"learning_rate": 9.918999696458006e-07,
"loss": 0.23538361489772797,
"step": 723,
"token_acc": 0.9311200744647844
},
{
"epoch": 0.15274261603375527,
"grad_norm": 0.64453125,
"learning_rate": 9.918696297413008e-07,
"loss": 0.2112676054239273,
"step": 724,
"token_acc": 0.9377406931964056
},
{
"epoch": 0.1529535864978903,
"grad_norm": 0.6171875,
"learning_rate": 9.918392335873457e-07,
"loss": 0.22141136229038239,
"step": 725,
"token_acc": 0.9383989145183175
},
{
"epoch": 0.15316455696202533,
"grad_norm": 0.8046875,
"learning_rate": 9.91808781187411e-07,
"loss": 0.27773499488830566,
"step": 726,
"token_acc": 0.9278820375335121
},
{
"epoch": 0.15337552742616034,
"grad_norm": 0.94921875,
"learning_rate": 9.917782725449799e-07,
"loss": 0.32096052169799805,
"step": 727,
"token_acc": 0.9175288205734555
},
{
"epoch": 0.15358649789029535,
"grad_norm": 0.71875,
"learning_rate": 9.91747707663541e-07,
"loss": 0.2451501190662384,
"step": 728,
"token_acc": 0.9271503803393798
},
{
"epoch": 0.15379746835443037,
"grad_norm": 1.546875,
"learning_rate": 9.917170865465894e-07,
"loss": 0.29911404848098755,
"step": 729,
"token_acc": 0.9180237372343362
},
{
"epoch": 0.1540084388185654,
"grad_norm": 0.953125,
"learning_rate": 9.91686409197627e-07,
"loss": 0.3019820749759674,
"step": 730,
"token_acc": 0.9172510518934082
},
{
"epoch": 0.15421940928270042,
"grad_norm": 1.1484375,
"learning_rate": 9.916556756201624e-07,
"loss": 0.281706839799881,
"step": 731,
"token_acc": 0.9272495213784301
},
{
"epoch": 0.15443037974683543,
"grad_norm": 0.93359375,
"learning_rate": 9.916248858177099e-07,
"loss": 0.27722233533859253,
"step": 732,
"token_acc": 0.9146039603960396
},
{
"epoch": 0.15464135021097047,
"grad_norm": 0.73046875,
"learning_rate": 9.915940397937906e-07,
"loss": 0.29295605421066284,
"step": 733,
"token_acc": 0.9225071225071225
},
{
"epoch": 0.1548523206751055,
"grad_norm": 0.80078125,
"learning_rate": 9.91563137551932e-07,
"loss": 0.24334131181240082,
"step": 734,
"token_acc": 0.9294417682062908
},
{
"epoch": 0.1550632911392405,
"grad_norm": 0.81640625,
"learning_rate": 9.91532179095668e-07,
"loss": 0.2572481334209442,
"step": 735,
"token_acc": 0.9311695579182988
},
{
"epoch": 0.15527426160337554,
"grad_norm": 0.75,
"learning_rate": 9.915011644285391e-07,
"loss": 0.26280131936073303,
"step": 736,
"token_acc": 0.9256795835743205
},
{
"epoch": 0.15548523206751055,
"grad_norm": 0.62890625,
"learning_rate": 9.91470093554092e-07,
"loss": 0.23156148195266724,
"step": 737,
"token_acc": 0.9377058999700509
},
{
"epoch": 0.15569620253164557,
"grad_norm": 0.60546875,
"learning_rate": 9.914389664758799e-07,
"loss": 0.24967870116233826,
"step": 738,
"token_acc": 0.9284906726964387
},
{
"epoch": 0.15590717299578058,
"grad_norm": 0.6484375,
"learning_rate": 9.914077831974626e-07,
"loss": 0.24829944968223572,
"step": 739,
"token_acc": 0.931044267877412
},
{
"epoch": 0.15611814345991562,
"grad_norm": 0.61328125,
"learning_rate": 9.91376543722406e-07,
"loss": 0.24180346727371216,
"step": 740,
"token_acc": 0.9371293001186239
},
{
"epoch": 0.15632911392405063,
"grad_norm": 0.84765625,
"learning_rate": 9.913452480542825e-07,
"loss": 0.26637858152389526,
"step": 741,
"token_acc": 0.9162153552086651
},
{
"epoch": 0.15654008438818565,
"grad_norm": 0.5625,
"learning_rate": 9.913138961966715e-07,
"loss": 0.22019389271736145,
"step": 742,
"token_acc": 0.9336415556159913
},
{
"epoch": 0.1567510548523207,
"grad_norm": 0.83203125,
"learning_rate": 9.912824881531577e-07,
"loss": 0.2972027361392975,
"step": 743,
"token_acc": 0.9233983286908078
},
{
"epoch": 0.1569620253164557,
"grad_norm": 0.6953125,
"learning_rate": 9.912510239273332e-07,
"loss": 0.26124250888824463,
"step": 744,
"token_acc": 0.9267277268942548
},
{
"epoch": 0.1571729957805907,
"grad_norm": 0.95703125,
"learning_rate": 9.912195035227964e-07,
"loss": 0.32723483443260193,
"step": 745,
"token_acc": 0.9195718654434251
},
{
"epoch": 0.15738396624472573,
"grad_norm": 0.71484375,
"learning_rate": 9.911879269431517e-07,
"loss": 0.23630690574645996,
"step": 746,
"token_acc": 0.9347326049453709
},
{
"epoch": 0.15759493670886077,
"grad_norm": 0.6796875,
"learning_rate": 9.911562941920099e-07,
"loss": 0.21784129738807678,
"step": 747,
"token_acc": 0.9337892446378614
},
{
"epoch": 0.15780590717299578,
"grad_norm": 0.76171875,
"learning_rate": 9.911246052729891e-07,
"loss": 0.26233282685279846,
"step": 748,
"token_acc": 0.9323260937991816
},
{
"epoch": 0.1580168776371308,
"grad_norm": 0.6328125,
"learning_rate": 9.910928601897126e-07,
"loss": 0.2327466756105423,
"step": 749,
"token_acc": 0.9362054681027341
},
{
"epoch": 0.15822784810126583,
"grad_norm": 0.7578125,
"learning_rate": 9.91061058945811e-07,
"loss": 0.27062827348709106,
"step": 750,
"token_acc": 0.918719909374115
},
{
"epoch": 0.15843881856540085,
"grad_norm": 0.671875,
"learning_rate": 9.910292015449211e-07,
"loss": 0.20303724706172943,
"step": 751,
"token_acc": 0.9412310547479121
},
{
"epoch": 0.15864978902953586,
"grad_norm": 0.65234375,
"learning_rate": 9.909972879906858e-07,
"loss": 0.24677664041519165,
"step": 752,
"token_acc": 0.925
},
{
"epoch": 0.15886075949367087,
"grad_norm": 0.7578125,
"learning_rate": 9.90965318286755e-07,
"loss": 0.2709593176841736,
"step": 753,
"token_acc": 0.9230769230769231
},
{
"epoch": 0.1590717299578059,
"grad_norm": 0.99609375,
"learning_rate": 9.909332924367846e-07,
"loss": 0.265384703874588,
"step": 754,
"token_acc": 0.9230769230769231
},
{
"epoch": 0.15928270042194093,
"grad_norm": 0.80078125,
"learning_rate": 9.909012104444368e-07,
"loss": 0.2868095636367798,
"step": 755,
"token_acc": 0.920038228735266
},
{
"epoch": 0.15949367088607594,
"grad_norm": 0.81640625,
"learning_rate": 9.908690723133807e-07,
"loss": 0.24986404180526733,
"step": 756,
"token_acc": 0.9256695756846224
},
{
"epoch": 0.15970464135021098,
"grad_norm": 0.62109375,
"learning_rate": 9.908368780472916e-07,
"loss": 0.20347082614898682,
"step": 757,
"token_acc": 0.9389263902282224
},
{
"epoch": 0.159915611814346,
"grad_norm": 0.59375,
"learning_rate": 9.908046276498511e-07,
"loss": 0.2612215578556061,
"step": 758,
"token_acc": 0.9279416235937975
},
{
"epoch": 0.160126582278481,
"grad_norm": 0.68359375,
"learning_rate": 9.907723211247472e-07,
"loss": 0.23647598922252655,
"step": 759,
"token_acc": 0.9354395604395604
},
{
"epoch": 0.16033755274261605,
"grad_norm": 0.72265625,
"learning_rate": 9.907399584756744e-07,
"loss": 0.28146815299987793,
"step": 760,
"token_acc": 0.92171219374824
},
{
"epoch": 0.16054852320675106,
"grad_norm": 0.72265625,
"learning_rate": 9.90707539706334e-07,
"loss": 0.2895510494709015,
"step": 761,
"token_acc": 0.9175170068027211
},
{
"epoch": 0.16075949367088607,
"grad_norm": 0.64453125,
"learning_rate": 9.90675064820433e-07,
"loss": 0.25281795859336853,
"step": 762,
"token_acc": 0.9276477832512315
},
{
"epoch": 0.16097046413502109,
"grad_norm": 0.67578125,
"learning_rate": 9.906425338216852e-07,
"loss": 0.2702397108078003,
"step": 763,
"token_acc": 0.9337220006136852
},
{
"epoch": 0.16118143459915613,
"grad_norm": 0.82421875,
"learning_rate": 9.906099467138111e-07,
"loss": 0.3201596736907959,
"step": 764,
"token_acc": 0.9149093599704032
},
{
"epoch": 0.16139240506329114,
"grad_norm": 0.8359375,
"learning_rate": 9.90577303500537e-07,
"loss": 0.26031017303466797,
"step": 765,
"token_acc": 0.922656699252444
},
{
"epoch": 0.16160337552742615,
"grad_norm": 0.84765625,
"learning_rate": 9.90544604185596e-07,
"loss": 0.2320261150598526,
"step": 766,
"token_acc": 0.933118216485773
},
{
"epoch": 0.1618143459915612,
"grad_norm": 0.69921875,
"learning_rate": 9.905118487727277e-07,
"loss": 0.2794190049171448,
"step": 767,
"token_acc": 0.9201467268623025
},
{
"epoch": 0.1620253164556962,
"grad_norm": 0.84375,
"learning_rate": 9.904790372656778e-07,
"loss": 0.2765384018421173,
"step": 768,
"token_acc": 0.9225014961101137
},
{
"epoch": 0.16223628691983122,
"grad_norm": 0.84765625,
"learning_rate": 9.904461696681984e-07,
"loss": 0.3068510890007019,
"step": 769,
"token_acc": 0.9177502267916541
},
{
"epoch": 0.16244725738396623,
"grad_norm": 0.71875,
"learning_rate": 9.904132459840485e-07,
"loss": 0.28465330600738525,
"step": 770,
"token_acc": 0.9240544629349471
},
{
"epoch": 0.16265822784810127,
"grad_norm": 0.75,
"learning_rate": 9.903802662169932e-07,
"loss": 0.2329617142677307,
"step": 771,
"token_acc": 0.9319875776397516
},
{
"epoch": 0.16286919831223629,
"grad_norm": 0.64453125,
"learning_rate": 9.903472303708038e-07,
"loss": 0.2284744679927826,
"step": 772,
"token_acc": 0.931237721021611
},
{
"epoch": 0.1630801687763713,
"grad_norm": 1.140625,
"learning_rate": 9.903141384492583e-07,
"loss": 0.23831237852573395,
"step": 773,
"token_acc": 0.9291455790413814
},
{
"epoch": 0.16329113924050634,
"grad_norm": 0.6796875,
"learning_rate": 9.902809904561414e-07,
"loss": 0.23870491981506348,
"step": 774,
"token_acc": 0.9348308374930671
},
{
"epoch": 0.16350210970464135,
"grad_norm": 0.75,
"learning_rate": 9.902477863952431e-07,
"loss": 0.27838945388793945,
"step": 775,
"token_acc": 0.9261146496815287
},
{
"epoch": 0.16371308016877636,
"grad_norm": 0.68359375,
"learning_rate": 9.902145262703613e-07,
"loss": 0.2492181956768036,
"step": 776,
"token_acc": 0.9293759512937595
},
{
"epoch": 0.1639240506329114,
"grad_norm": 0.58984375,
"learning_rate": 9.901812100852993e-07,
"loss": 0.2085292637348175,
"step": 777,
"token_acc": 0.9389517569982132
},
{
"epoch": 0.16413502109704642,
"grad_norm": 0.953125,
"learning_rate": 9.90147837843867e-07,
"loss": 0.2694481909275055,
"step": 778,
"token_acc": 0.9280432309442548
},
{
"epoch": 0.16434599156118143,
"grad_norm": 0.80859375,
"learning_rate": 9.901144095498808e-07,
"loss": 0.25209715962409973,
"step": 779,
"token_acc": 0.9302030456852792
},
{
"epoch": 0.16455696202531644,
"grad_norm": 0.88671875,
"learning_rate": 9.900809252071635e-07,
"loss": 0.31358861923217773,
"step": 780,
"token_acc": 0.9151069518716578
},
{
"epoch": 0.16476793248945149,
"grad_norm": 0.734375,
"learning_rate": 9.900473848195446e-07,
"loss": 0.23959940671920776,
"step": 781,
"token_acc": 0.9326456310679612
},
{
"epoch": 0.1649789029535865,
"grad_norm": 0.80078125,
"learning_rate": 9.900137883908592e-07,
"loss": 0.29789382219314575,
"step": 782,
"token_acc": 0.9166666666666666
},
{
"epoch": 0.1651898734177215,
"grad_norm": 0.796875,
"learning_rate": 9.8998013592495e-07,
"loss": 0.22469905018806458,
"step": 783,
"token_acc": 0.933932193567082
},
{
"epoch": 0.16540084388185655,
"grad_norm": 0.6875,
"learning_rate": 9.89946427425665e-07,
"loss": 0.28561171889305115,
"step": 784,
"token_acc": 0.9251565167899829
},
{
"epoch": 0.16561181434599156,
"grad_norm": 0.90625,
"learning_rate": 9.89912662896859e-07,
"loss": 0.28935301303863525,
"step": 785,
"token_acc": 0.9215164615896242
},
{
"epoch": 0.16582278481012658,
"grad_norm": 0.73046875,
"learning_rate": 9.898788423423935e-07,
"loss": 0.2708919048309326,
"step": 786,
"token_acc": 0.927591706539075
},
{
"epoch": 0.1660337552742616,
"grad_norm": 0.66015625,
"learning_rate": 9.898449657661362e-07,
"loss": 0.2672666311264038,
"step": 787,
"token_acc": 0.9263598326359833
},
{
"epoch": 0.16624472573839663,
"grad_norm": 0.8046875,
"learning_rate": 9.89811033171961e-07,
"loss": 0.2862321734428406,
"step": 788,
"token_acc": 0.9180470793374019
},
{
"epoch": 0.16645569620253164,
"grad_norm": 0.8046875,
"learning_rate": 9.897770445637483e-07,
"loss": 0.2871711850166321,
"step": 789,
"token_acc": 0.9249183895538629
},
{
"epoch": 0.16666666666666666,
"grad_norm": 0.80859375,
"learning_rate": 9.897429999453852e-07,
"loss": 0.2397966980934143,
"step": 790,
"token_acc": 0.9377081945369754
},
{
"epoch": 0.1668776371308017,
"grad_norm": 0.80859375,
"learning_rate": 9.89708899320765e-07,
"loss": 0.2998412847518921,
"step": 791,
"token_acc": 0.9202363367799113
},
{
"epoch": 0.1670886075949367,
"grad_norm": 0.6640625,
"learning_rate": 9.89674742693787e-07,
"loss": 0.27035748958587646,
"step": 792,
"token_acc": 0.9229891614375356
},
{
"epoch": 0.16729957805907172,
"grad_norm": 0.76953125,
"learning_rate": 9.89640530068358e-07,
"loss": 0.17933598160743713,
"step": 793,
"token_acc": 0.9483188044831881
},
{
"epoch": 0.16751054852320676,
"grad_norm": 0.91796875,
"learning_rate": 9.896062614483898e-07,
"loss": 0.2540227472782135,
"step": 794,
"token_acc": 0.929927414852038
},
{
"epoch": 0.16772151898734178,
"grad_norm": 0.68359375,
"learning_rate": 9.895719368378016e-07,
"loss": 0.24861711263656616,
"step": 795,
"token_acc": 0.9333129397369226
},
{
"epoch": 0.1679324894514768,
"grad_norm": 0.69140625,
"learning_rate": 9.89537556240519e-07,
"loss": 0.2633477449417114,
"step": 796,
"token_acc": 0.9196560924992588
},
{
"epoch": 0.1681434599156118,
"grad_norm": 0.8671875,
"learning_rate": 9.89503119660473e-07,
"loss": 0.2524856925010681,
"step": 797,
"token_acc": 0.9313725490196079
},
{
"epoch": 0.16835443037974684,
"grad_norm": 0.91796875,
"learning_rate": 9.894686271016027e-07,
"loss": 0.30388563871383667,
"step": 798,
"token_acc": 0.9198352344740177
},
{
"epoch": 0.16856540084388186,
"grad_norm": 0.6328125,
"learning_rate": 9.894340785678517e-07,
"loss": 0.2910333573818207,
"step": 799,
"token_acc": 0.9234449760765551
},
{
"epoch": 0.16877637130801687,
"grad_norm": 0.859375,
"learning_rate": 9.893994740631713e-07,
"loss": 0.25983309745788574,
"step": 800,
"token_acc": 0.9273416982783775
},
{
"epoch": 0.16877637130801687,
"eval_loss": 0.43369975686073303,
"eval_runtime": 245.7245,
"eval_samples_per_second": 137.166,
"eval_steps_per_second": 2.145,
"eval_token_acc": 0.8991631517544647,
"step": 800
},
{
"epoch": 0.1689873417721519,
"grad_norm": 0.640625,
"learning_rate": 9.893648135915188e-07,
"loss": 0.26705414056777954,
"step": 801,
"token_acc": 0.9297355062783863
},
{
"epoch": 0.16919831223628692,
"grad_norm": 0.68359375,
"learning_rate": 9.893300971568578e-07,
"loss": 0.2386769950389862,
"step": 802,
"token_acc": 0.935367545076283
},
{
"epoch": 0.16940928270042194,
"grad_norm": 0.71875,
"learning_rate": 9.892953247631589e-07,
"loss": 0.2654857337474823,
"step": 803,
"token_acc": 0.9260304912478825
},
{
"epoch": 0.16962025316455695,
"grad_norm": 0.71484375,
"learning_rate": 9.89260496414398e-07,
"loss": 0.2585882842540741,
"step": 804,
"token_acc": 0.9277403551745255
},
{
"epoch": 0.169831223628692,
"grad_norm": 0.76953125,
"learning_rate": 9.892256121145584e-07,
"loss": 0.25679513812065125,
"step": 805,
"token_acc": 0.928305133352452
},
{
"epoch": 0.170042194092827,
"grad_norm": 0.671875,
"learning_rate": 9.891906718676291e-07,
"loss": 0.269248902797699,
"step": 806,
"token_acc": 0.9328318108543794
},
{
"epoch": 0.17025316455696202,
"grad_norm": 0.69140625,
"learning_rate": 9.89155675677606e-07,
"loss": 0.2557290196418762,
"step": 807,
"token_acc": 0.9275456919060052
},
{
"epoch": 0.17046413502109706,
"grad_norm": 0.63671875,
"learning_rate": 9.891206235484913e-07,
"loss": 0.23980513215065002,
"step": 808,
"token_acc": 0.9353355807539074
},
{
"epoch": 0.17067510548523207,
"grad_norm": 0.65625,
"learning_rate": 9.890855154842935e-07,
"loss": 0.2392064481973648,
"step": 809,
"token_acc": 0.9303870595031773
},
{
"epoch": 0.17088607594936708,
"grad_norm": 0.7890625,
"learning_rate": 9.890503514890275e-07,
"loss": 0.23739401996135712,
"step": 810,
"token_acc": 0.9328014728444308
},
{
"epoch": 0.1710970464135021,
"grad_norm": 0.7890625,
"learning_rate": 9.89015131566714e-07,
"loss": 0.2900955080986023,
"step": 811,
"token_acc": 0.9233797698364627
},
{
"epoch": 0.17130801687763714,
"grad_norm": 0.69921875,
"learning_rate": 9.889798557213818e-07,
"loss": 0.27924585342407227,
"step": 812,
"token_acc": 0.9214012363850457
},
{
"epoch": 0.17151898734177215,
"grad_norm": 0.7421875,
"learning_rate": 9.88944523957064e-07,
"loss": 0.2851244807243347,
"step": 813,
"token_acc": 0.9264660254421346
},
{
"epoch": 0.17172995780590716,
"grad_norm": 1.1875,
"learning_rate": 9.889091362778017e-07,
"loss": 0.23967793583869934,
"step": 814,
"token_acc": 0.9278113316077078
},
{
"epoch": 0.1719409282700422,
"grad_norm": 0.6796875,
"learning_rate": 9.888736926876415e-07,
"loss": 0.23629070818424225,
"step": 815,
"token_acc": 0.9284259528658714
},
{
"epoch": 0.17215189873417722,
"grad_norm": 0.93359375,
"learning_rate": 9.88838193190637e-07,
"loss": 0.3324328064918518,
"step": 816,
"token_acc": 0.9166411277965063
},
{
"epoch": 0.17236286919831223,
"grad_norm": 0.73046875,
"learning_rate": 9.888026377908472e-07,
"loss": 0.2603840231895447,
"step": 817,
"token_acc": 0.9286516853932584
},
{
"epoch": 0.17257383966244727,
"grad_norm": 0.7421875,
"learning_rate": 9.887670264923387e-07,
"loss": 0.25500792264938354,
"step": 818,
"token_acc": 0.9334559950935296
},
{
"epoch": 0.17278481012658228,
"grad_norm": 0.69921875,
"learning_rate": 9.88731359299184e-07,
"loss": 0.25971531867980957,
"step": 819,
"token_acc": 0.928698752228164
},
{
"epoch": 0.1729957805907173,
"grad_norm": 0.65625,
"learning_rate": 9.886956362154617e-07,
"loss": 0.2521659731864929,
"step": 820,
"token_acc": 0.9252336448598131
},
{
"epoch": 0.1732067510548523,
"grad_norm": 1.0078125,
"learning_rate": 9.88659857245257e-07,
"loss": 0.23598764836788177,
"step": 821,
"token_acc": 0.9364791288566243
},
{
"epoch": 0.17341772151898735,
"grad_norm": 2.28125,
"learning_rate": 9.886240223926617e-07,
"loss": 0.2466164529323578,
"step": 822,
"token_acc": 0.9289433384379786
},
{
"epoch": 0.17362869198312236,
"grad_norm": 0.609375,
"learning_rate": 9.88588131661774e-07,
"loss": 0.2287391722202301,
"step": 823,
"token_acc": 0.9339788732394366
},
{
"epoch": 0.17383966244725738,
"grad_norm": 0.9921875,
"learning_rate": 9.885521850566977e-07,
"loss": 0.3011782765388489,
"step": 824,
"token_acc": 0.9158669225847729
},
{
"epoch": 0.17405063291139242,
"grad_norm": 0.88671875,
"learning_rate": 9.88516182581544e-07,
"loss": 0.24732713401317596,
"step": 825,
"token_acc": 0.9297990096125838
},
{
"epoch": 0.17426160337552743,
"grad_norm": 0.8046875,
"learning_rate": 9.884801242404303e-07,
"loss": 0.2557525038719177,
"step": 826,
"token_acc": 0.9272430668841762
},
{
"epoch": 0.17447257383966244,
"grad_norm": 0.58203125,
"learning_rate": 9.884440100374798e-07,
"loss": 0.2030971348285675,
"step": 827,
"token_acc": 0.9397905759162304
},
{
"epoch": 0.17468354430379746,
"grad_norm": 0.72265625,
"learning_rate": 9.884078399768226e-07,
"loss": 0.23457200825214386,
"step": 828,
"token_acc": 0.9320594479830149
},
{
"epoch": 0.1748945147679325,
"grad_norm": 0.75,
"learning_rate": 9.88371614062595e-07,
"loss": 0.27027198672294617,
"step": 829,
"token_acc": 0.9281524926686217
},
{
"epoch": 0.1751054852320675,
"grad_norm": 0.578125,
"learning_rate": 9.8833533229894e-07,
"loss": 0.2117438018321991,
"step": 830,
"token_acc": 0.9373803664205633
},
{
"epoch": 0.17531645569620252,
"grad_norm": 0.71484375,
"learning_rate": 9.882989946900063e-07,
"loss": 0.21879255771636963,
"step": 831,
"token_acc": 0.9381270903010034
},
{
"epoch": 0.17552742616033756,
"grad_norm": 0.75,
"learning_rate": 9.882626012399495e-07,
"loss": 0.27527743577957153,
"step": 832,
"token_acc": 0.9288235294117647
},
{
"epoch": 0.17573839662447258,
"grad_norm": 0.74609375,
"learning_rate": 9.882261519529318e-07,
"loss": 0.2788648307323456,
"step": 833,
"token_acc": 0.9233587786259542
},
{
"epoch": 0.1759493670886076,
"grad_norm": 0.75,
"learning_rate": 9.881896468331215e-07,
"loss": 0.26945775747299194,
"step": 834,
"token_acc": 0.9267376330619912
},
{
"epoch": 0.17616033755274263,
"grad_norm": 0.89453125,
"learning_rate": 9.881530858846928e-07,
"loss": 0.30408790707588196,
"step": 835,
"token_acc": 0.9146719234018587
},
{
"epoch": 0.17637130801687764,
"grad_norm": 0.8125,
"learning_rate": 9.88116469111827e-07,
"loss": 0.2584350109100342,
"step": 836,
"token_acc": 0.9246458923512748
},
{
"epoch": 0.17658227848101266,
"grad_norm": 0.8125,
"learning_rate": 9.880797965187119e-07,
"loss": 0.3014131784439087,
"step": 837,
"token_acc": 0.9151281344723065
},
{
"epoch": 0.17679324894514767,
"grad_norm": 0.59375,
"learning_rate": 9.880430681095407e-07,
"loss": 0.21773123741149902,
"step": 838,
"token_acc": 0.9367166004280036
},
{
"epoch": 0.1770042194092827,
"grad_norm": 1.9609375,
"learning_rate": 9.88006283888514e-07,
"loss": 0.26315808296203613,
"step": 839,
"token_acc": 0.9259569712210115
},
{
"epoch": 0.17721518987341772,
"grad_norm": 0.93359375,
"learning_rate": 9.879694438598383e-07,
"loss": 0.2646620571613312,
"step": 840,
"token_acc": 0.9318522966076254
},
{
"epoch": 0.17742616033755274,
"grad_norm": 0.78125,
"learning_rate": 9.879325480277266e-07,
"loss": 0.2713755965232849,
"step": 841,
"token_acc": 0.9206214689265537
},
{
"epoch": 0.17763713080168778,
"grad_norm": 0.859375,
"learning_rate": 9.878955963963979e-07,
"loss": 0.26667535305023193,
"step": 842,
"token_acc": 0.9262686567164179
},
{
"epoch": 0.1778481012658228,
"grad_norm": 0.734375,
"learning_rate": 9.878585889700785e-07,
"loss": 0.24986431002616882,
"step": 843,
"token_acc": 0.9370695053224797
},
{
"epoch": 0.1780590717299578,
"grad_norm": 1.4140625,
"learning_rate": 9.878215257530004e-07,
"loss": 0.2651556730270386,
"step": 844,
"token_acc": 0.9286946520989074
},
{
"epoch": 0.17827004219409281,
"grad_norm": 0.828125,
"learning_rate": 9.877844067494017e-07,
"loss": 0.2608075737953186,
"step": 845,
"token_acc": 0.9263001485884101
},
{
"epoch": 0.17848101265822786,
"grad_norm": 0.75390625,
"learning_rate": 9.877472319635275e-07,
"loss": 0.28958860039711,
"step": 846,
"token_acc": 0.9233128834355828
},
{
"epoch": 0.17869198312236287,
"grad_norm": 1.0234375,
"learning_rate": 9.877100013996291e-07,
"loss": 0.2941049039363861,
"step": 847,
"token_acc": 0.9213372664700098
},
{
"epoch": 0.17890295358649788,
"grad_norm": 0.84375,
"learning_rate": 9.876727150619642e-07,
"loss": 0.2620714604854584,
"step": 848,
"token_acc": 0.9288433382137629
},
{
"epoch": 0.17911392405063292,
"grad_norm": 0.60546875,
"learning_rate": 9.876353729547968e-07,
"loss": 0.2020449936389923,
"step": 849,
"token_acc": 0.9392366412213741
},
{
"epoch": 0.17932489451476794,
"grad_norm": 0.953125,
"learning_rate": 9.875979750823969e-07,
"loss": 0.2892880439758301,
"step": 850,
"token_acc": 0.9246448424953675
},
{
"epoch": 0.17953586497890295,
"grad_norm": 0.70703125,
"learning_rate": 9.875605214490417e-07,
"loss": 0.2778629660606384,
"step": 851,
"token_acc": 0.923582580115037
},
{
"epoch": 0.17974683544303796,
"grad_norm": 0.5859375,
"learning_rate": 9.875230120590142e-07,
"loss": 0.23370903730392456,
"step": 852,
"token_acc": 0.9249401117913228
},
{
"epoch": 0.179957805907173,
"grad_norm": 0.75390625,
"learning_rate": 9.874854469166038e-07,
"loss": 0.28334856033325195,
"step": 853,
"token_acc": 0.9264705882352942
},
{
"epoch": 0.18016877637130801,
"grad_norm": 0.55078125,
"learning_rate": 9.874478260261067e-07,
"loss": 0.2282511293888092,
"step": 854,
"token_acc": 0.933620159803319
},
{
"epoch": 0.18037974683544303,
"grad_norm": 1.1015625,
"learning_rate": 9.874101493918249e-07,
"loss": 0.27366286516189575,
"step": 855,
"token_acc": 0.9260089686098655
},
{
"epoch": 0.18059071729957807,
"grad_norm": 0.6875,
"learning_rate": 9.87372417018067e-07,
"loss": 0.25619056820869446,
"step": 856,
"token_acc": 0.9255610290093049
},
{
"epoch": 0.18080168776371308,
"grad_norm": 0.828125,
"learning_rate": 9.873346289091483e-07,
"loss": 0.270757257938385,
"step": 857,
"token_acc": 0.923998738568275
},
{
"epoch": 0.1810126582278481,
"grad_norm": 0.66796875,
"learning_rate": 9.8729678506939e-07,
"loss": 0.28628918528556824,
"step": 858,
"token_acc": 0.9234957020057306
},
{
"epoch": 0.18122362869198314,
"grad_norm": 0.7265625,
"learning_rate": 9.872588855031197e-07,
"loss": 0.2525092661380768,
"step": 859,
"token_acc": 0.9317073170731708
},
{
"epoch": 0.18143459915611815,
"grad_norm": 0.83984375,
"learning_rate": 9.872209302146718e-07,
"loss": 0.28244319558143616,
"step": 860,
"token_acc": 0.9267202859696158
},
{
"epoch": 0.18164556962025316,
"grad_norm": 0.828125,
"learning_rate": 9.871829192083867e-07,
"loss": 0.254133403301239,
"step": 861,
"token_acc": 0.928436911487759
},
{
"epoch": 0.18185654008438817,
"grad_norm": 0.75,
"learning_rate": 9.871448524886113e-07,
"loss": 0.2619815468788147,
"step": 862,
"token_acc": 0.9243888573052871
},
{
"epoch": 0.18206751054852321,
"grad_norm": 0.91796875,
"learning_rate": 9.87106730059699e-07,
"loss": 0.2682092487812042,
"step": 863,
"token_acc": 0.9261158021712907
},
{
"epoch": 0.18227848101265823,
"grad_norm": 0.6015625,
"learning_rate": 9.870685519260092e-07,
"loss": 0.245108962059021,
"step": 864,
"token_acc": 0.9272880404267265
},
{
"epoch": 0.18248945147679324,
"grad_norm": 0.9609375,
"learning_rate": 9.870303180919078e-07,
"loss": 0.2907876670360565,
"step": 865,
"token_acc": 0.9245337159253946
},
{
"epoch": 0.18270042194092828,
"grad_norm": 0.60546875,
"learning_rate": 9.869920285617676e-07,
"loss": 0.24249601364135742,
"step": 866,
"token_acc": 0.931304347826087
},
{
"epoch": 0.1829113924050633,
"grad_norm": 0.83203125,
"learning_rate": 9.869536833399669e-07,
"loss": 0.2370653748512268,
"step": 867,
"token_acc": 0.9323520200438459
},
{
"epoch": 0.1831223628691983,
"grad_norm": 0.859375,
"learning_rate": 9.869152824308912e-07,
"loss": 0.3008883595466614,
"step": 868,
"token_acc": 0.9213197969543148
},
{
"epoch": 0.18333333333333332,
"grad_norm": 0.8125,
"learning_rate": 9.868768258389314e-07,
"loss": 0.2317754030227661,
"step": 869,
"token_acc": 0.9351635514018691
},
{
"epoch": 0.18354430379746836,
"grad_norm": 0.71484375,
"learning_rate": 9.868383135684857e-07,
"loss": 0.2313736081123352,
"step": 870,
"token_acc": 0.9373202990224266
},
{
"epoch": 0.18375527426160337,
"grad_norm": 0.9296875,
"learning_rate": 9.867997456239586e-07,
"loss": 0.28026607632637024,
"step": 871,
"token_acc": 0.9176308539944904
},
{
"epoch": 0.1839662447257384,
"grad_norm": 0.6875,
"learning_rate": 9.8676112200976e-07,
"loss": 0.254774272441864,
"step": 872,
"token_acc": 0.9306763962952568
},
{
"epoch": 0.18417721518987343,
"grad_norm": 0.75,
"learning_rate": 9.867224427303073e-07,
"loss": 0.24183842539787292,
"step": 873,
"token_acc": 0.9341463414634147
},
{
"epoch": 0.18438818565400844,
"grad_norm": 0.86328125,
"learning_rate": 9.86683707790024e-07,
"loss": 0.23453059792518616,
"step": 874,
"token_acc": 0.9323812299621101
},
{
"epoch": 0.18459915611814345,
"grad_norm": 0.734375,
"learning_rate": 9.86644917193339e-07,
"loss": 0.24839141964912415,
"step": 875,
"token_acc": 0.9287037037037037
},
{
"epoch": 0.1848101265822785,
"grad_norm": 0.671875,
"learning_rate": 9.86606070944689e-07,
"loss": 0.2521136403083801,
"step": 876,
"token_acc": 0.9300189993666877
},
{
"epoch": 0.1850210970464135,
"grad_norm": 1.046875,
"learning_rate": 9.865671690485162e-07,
"loss": 0.3050832748413086,
"step": 877,
"token_acc": 0.9174647887323943
},
{
"epoch": 0.18523206751054852,
"grad_norm": 0.74609375,
"learning_rate": 9.865282115092692e-07,
"loss": 0.2835577726364136,
"step": 878,
"token_acc": 0.9225286643941741
},
{
"epoch": 0.18544303797468353,
"grad_norm": 0.7890625,
"learning_rate": 9.864891983314033e-07,
"loss": 0.29184651374816895,
"step": 879,
"token_acc": 0.9218231210383339
},
{
"epoch": 0.18565400843881857,
"grad_norm": 2.859375,
"learning_rate": 9.8645012951938e-07,
"loss": 0.2807004451751709,
"step": 880,
"token_acc": 0.9238008500303583
},
{
"epoch": 0.1858649789029536,
"grad_norm": 0.62109375,
"learning_rate": 9.864110050776672e-07,
"loss": 0.25495046377182007,
"step": 881,
"token_acc": 0.9281183932346723
},
{
"epoch": 0.1860759493670886,
"grad_norm": 0.62890625,
"learning_rate": 9.86371825010739e-07,
"loss": 0.26357853412628174,
"step": 882,
"token_acc": 0.9274457329765091
},
{
"epoch": 0.18628691983122364,
"grad_norm": 0.703125,
"learning_rate": 9.86332589323076e-07,
"loss": 0.2856602966785431,
"step": 883,
"token_acc": 0.9247496423462088
},
{
"epoch": 0.18649789029535865,
"grad_norm": 0.63671875,
"learning_rate": 9.862932980191652e-07,
"loss": 0.26217591762542725,
"step": 884,
"token_acc": 0.9308156140907649
},
{
"epoch": 0.18670886075949367,
"grad_norm": 0.859375,
"learning_rate": 9.862539511034997e-07,
"loss": 0.2957126498222351,
"step": 885,
"token_acc": 0.9175007582650895
},
{
"epoch": 0.18691983122362868,
"grad_norm": 0.71875,
"learning_rate": 9.862145485805793e-07,
"loss": 0.2381889373064041,
"step": 886,
"token_acc": 0.9338040600176523
},
{
"epoch": 0.18713080168776372,
"grad_norm": 0.61328125,
"learning_rate": 9.861750904549099e-07,
"loss": 0.23038305342197418,
"step": 887,
"token_acc": 0.933588010578901
},
{
"epoch": 0.18734177215189873,
"grad_norm": 0.7578125,
"learning_rate": 9.86135576731004e-07,
"loss": 0.2670343518257141,
"step": 888,
"token_acc": 0.9209346991037132
},
{
"epoch": 0.18755274261603375,
"grad_norm": 0.875,
"learning_rate": 9.860960074133802e-07,
"loss": 0.3037135899066925,
"step": 889,
"token_acc": 0.9173450219160927
},
{
"epoch": 0.1877637130801688,
"grad_norm": 0.6953125,
"learning_rate": 9.860563825065637e-07,
"loss": 0.23587052524089813,
"step": 890,
"token_acc": 0.9326950971859588
},
{
"epoch": 0.1879746835443038,
"grad_norm": 1.046875,
"learning_rate": 9.86016702015086e-07,
"loss": 0.2837037444114685,
"step": 891,
"token_acc": 0.9186681222707423
},
{
"epoch": 0.1881856540084388,
"grad_norm": 0.80078125,
"learning_rate": 9.85976965943485e-07,
"loss": 0.273685485124588,
"step": 892,
"token_acc": 0.9198871650211565
},
{
"epoch": 0.18839662447257383,
"grad_norm": 0.74609375,
"learning_rate": 9.859371742963043e-07,
"loss": 0.24621078372001648,
"step": 893,
"token_acc": 0.9370564640543042
},
{
"epoch": 0.18860759493670887,
"grad_norm": 0.63671875,
"learning_rate": 9.85897327078095e-07,
"loss": 0.2355155646800995,
"step": 894,
"token_acc": 0.933295647258338
},
{
"epoch": 0.18881856540084388,
"grad_norm": 0.6875,
"learning_rate": 9.858574242934136e-07,
"loss": 0.29725679755210876,
"step": 895,
"token_acc": 0.9205632306057385
},
{
"epoch": 0.1890295358649789,
"grad_norm": 0.71484375,
"learning_rate": 9.858174659468237e-07,
"loss": 0.23919257521629333,
"step": 896,
"token_acc": 0.9367167919799498
},
{
"epoch": 0.18924050632911393,
"grad_norm": 0.6640625,
"learning_rate": 9.857774520428945e-07,
"loss": 0.2421645075082779,
"step": 897,
"token_acc": 0.928311057108141
},
{
"epoch": 0.18945147679324895,
"grad_norm": 0.90625,
"learning_rate": 9.85737382586202e-07,
"loss": 0.22805655002593994,
"step": 898,
"token_acc": 0.9353342428376534
},
{
"epoch": 0.18966244725738396,
"grad_norm": 1.125,
"learning_rate": 9.856972575813285e-07,
"loss": 0.2736568748950958,
"step": 899,
"token_acc": 0.9199507389162561
},
{
"epoch": 0.189873417721519,
"grad_norm": 0.71875,
"learning_rate": 9.85657077032863e-07,
"loss": 0.251034677028656,
"step": 900,
"token_acc": 0.933903806432576
},
{
"epoch": 0.190084388185654,
"grad_norm": 0.671875,
"learning_rate": 9.856168409454e-07,
"loss": 0.2377174347639084,
"step": 901,
"token_acc": 0.9313658201784488
},
{
"epoch": 0.19029535864978903,
"grad_norm": 0.625,
"learning_rate": 9.855765493235408e-07,
"loss": 0.27164188027381897,
"step": 902,
"token_acc": 0.9263128176171654
},
{
"epoch": 0.19050632911392404,
"grad_norm": 0.84765625,
"learning_rate": 9.855362021718936e-07,
"loss": 0.250331312417984,
"step": 903,
"token_acc": 0.9330877839165131
},
{
"epoch": 0.19071729957805908,
"grad_norm": 0.60546875,
"learning_rate": 9.85495799495072e-07,
"loss": 0.20965948700904846,
"step": 904,
"token_acc": 0.936726272352132
},
{
"epoch": 0.1909282700421941,
"grad_norm": 0.6640625,
"learning_rate": 9.854553412976965e-07,
"loss": 0.24447084963321686,
"step": 905,
"token_acc": 0.9254159495123351
},
{
"epoch": 0.1911392405063291,
"grad_norm": 0.6875,
"learning_rate": 9.854148275843939e-07,
"loss": 0.2490314543247223,
"step": 906,
"token_acc": 0.9285930408472012
},
{
"epoch": 0.19135021097046415,
"grad_norm": 0.61328125,
"learning_rate": 9.853742583597973e-07,
"loss": 0.21816563606262207,
"step": 907,
"token_acc": 0.9387067116150781
},
{
"epoch": 0.19156118143459916,
"grad_norm": 0.66015625,
"learning_rate": 9.853336336285461e-07,
"loss": 0.24077807366847992,
"step": 908,
"token_acc": 0.9278679026651216
},
{
"epoch": 0.19177215189873417,
"grad_norm": 0.82421875,
"learning_rate": 9.852929533952858e-07,
"loss": 0.2617112696170807,
"step": 909,
"token_acc": 0.9296264118158123
},
{
"epoch": 0.19198312236286919,
"grad_norm": 0.75,
"learning_rate": 9.852522176646692e-07,
"loss": 0.22484534978866577,
"step": 910,
"token_acc": 0.9368040926873308
},
{
"epoch": 0.19219409282700423,
"grad_norm": 0.671875,
"learning_rate": 9.85211426441354e-07,
"loss": 0.2532415986061096,
"step": 911,
"token_acc": 0.9334488734835356
},
{
"epoch": 0.19240506329113924,
"grad_norm": 0.87890625,
"learning_rate": 9.851705797300056e-07,
"loss": 0.31424853205680847,
"step": 912,
"token_acc": 0.9197608558842039
},
{
"epoch": 0.19261603375527425,
"grad_norm": 0.70703125,
"learning_rate": 9.851296775352948e-07,
"loss": 0.29285135865211487,
"step": 913,
"token_acc": 0.9182986536107711
},
{
"epoch": 0.1928270042194093,
"grad_norm": 0.59765625,
"learning_rate": 9.850887198618996e-07,
"loss": 0.21450576186180115,
"step": 914,
"token_acc": 0.9373626373626374
},
{
"epoch": 0.1930379746835443,
"grad_norm": 0.78515625,
"learning_rate": 9.850477067145031e-07,
"loss": 0.2844701111316681,
"step": 915,
"token_acc": 0.919885094158953
},
{
"epoch": 0.19324894514767932,
"grad_norm": 0.68359375,
"learning_rate": 9.850066380977961e-07,
"loss": 0.26549211144447327,
"step": 916,
"token_acc": 0.9245901639344263
},
{
"epoch": 0.19345991561181436,
"grad_norm": 0.765625,
"learning_rate": 9.849655140164752e-07,
"loss": 0.258350133895874,
"step": 917,
"token_acc": 0.918646080760095
},
{
"epoch": 0.19367088607594937,
"grad_norm": 0.7421875,
"learning_rate": 9.849243344752427e-07,
"loss": 0.2719504237174988,
"step": 918,
"token_acc": 0.9267654751525719
},
{
"epoch": 0.19388185654008439,
"grad_norm": 0.79296875,
"learning_rate": 9.848830994788083e-07,
"loss": 0.27195435762405396,
"step": 919,
"token_acc": 0.9220445459737293
},
{
"epoch": 0.1940928270042194,
"grad_norm": 0.71484375,
"learning_rate": 9.848418090318876e-07,
"loss": 0.24952857196331024,
"step": 920,
"token_acc": 0.9340033500837521
},
{
"epoch": 0.19430379746835444,
"grad_norm": 0.7421875,
"learning_rate": 9.848004631392022e-07,
"loss": 0.22502082586288452,
"step": 921,
"token_acc": 0.935454267360049
},
{
"epoch": 0.19451476793248945,
"grad_norm": 0.76171875,
"learning_rate": 9.847590618054806e-07,
"loss": 0.30236607789993286,
"step": 922,
"token_acc": 0.9160954208938854
},
{
"epoch": 0.19472573839662446,
"grad_norm": 0.7578125,
"learning_rate": 9.847176050354573e-07,
"loss": 0.26875466108322144,
"step": 923,
"token_acc": 0.9231622746185852
},
{
"epoch": 0.1949367088607595,
"grad_norm": 0.79296875,
"learning_rate": 9.846760928338734e-07,
"loss": 0.21099932491779327,
"step": 924,
"token_acc": 0.9381474710542352
},
{
"epoch": 0.19514767932489452,
"grad_norm": 2.921875,
"learning_rate": 9.846345252054758e-07,
"loss": 0.24902689456939697,
"step": 925,
"token_acc": 0.9338211899459116
},
{
"epoch": 0.19535864978902953,
"grad_norm": 0.80859375,
"learning_rate": 9.845929021550184e-07,
"loss": 0.22670647501945496,
"step": 926,
"token_acc": 0.9382314694408322
},
{
"epoch": 0.19556962025316454,
"grad_norm": 0.70703125,
"learning_rate": 9.84551223687261e-07,
"loss": 0.24977506697177887,
"step": 927,
"token_acc": 0.927700089259149
},
{
"epoch": 0.19578059071729959,
"grad_norm": 0.7109375,
"learning_rate": 9.8450948980697e-07,
"loss": 0.2701714038848877,
"step": 928,
"token_acc": 0.9202069716775599
},
{
"epoch": 0.1959915611814346,
"grad_norm": 0.82421875,
"learning_rate": 9.844677005189182e-07,
"loss": 0.2738378643989563,
"step": 929,
"token_acc": 0.9230093676814989
},
{
"epoch": 0.1962025316455696,
"grad_norm": 0.7890625,
"learning_rate": 9.844258558278842e-07,
"loss": 0.2802038788795471,
"step": 930,
"token_acc": 0.9218846869187849
},
{
"epoch": 0.19641350210970465,
"grad_norm": 1.125,
"learning_rate": 9.843839557386534e-07,
"loss": 0.28460338711738586,
"step": 931,
"token_acc": 0.9176701204144497
},
{
"epoch": 0.19662447257383966,
"grad_norm": 0.70703125,
"learning_rate": 9.843420002560173e-07,
"loss": 0.2364983856678009,
"step": 932,
"token_acc": 0.9339788732394366
},
{
"epoch": 0.19683544303797468,
"grad_norm": 0.69921875,
"learning_rate": 9.842999893847744e-07,
"loss": 0.24972565472126007,
"step": 933,
"token_acc": 0.9282414536495226
},
{
"epoch": 0.19704641350210972,
"grad_norm": 0.62890625,
"learning_rate": 9.842579231297284e-07,
"loss": 0.23772844672203064,
"step": 934,
"token_acc": 0.9301578024547048
},
{
"epoch": 0.19725738396624473,
"grad_norm": 0.86328125,
"learning_rate": 9.842158014956901e-07,
"loss": 0.2724204659461975,
"step": 935,
"token_acc": 0.920952380952381
},
{
"epoch": 0.19746835443037974,
"grad_norm": 0.67578125,
"learning_rate": 9.841736244874769e-07,
"loss": 0.20951035618782043,
"step": 936,
"token_acc": 0.9367752622860298
},
{
"epoch": 0.19767932489451476,
"grad_norm": 1.84375,
"learning_rate": 9.841313921099112e-07,
"loss": 0.2654408812522888,
"step": 937,
"token_acc": 0.9252709640616087
},
{
"epoch": 0.1978902953586498,
"grad_norm": 0.609375,
"learning_rate": 9.840891043678235e-07,
"loss": 0.20615626871585846,
"step": 938,
"token_acc": 0.9382829208677055
},
{
"epoch": 0.1981012658227848,
"grad_norm": 0.78515625,
"learning_rate": 9.840467612660494e-07,
"loss": 0.24997380375862122,
"step": 939,
"token_acc": 0.9305555555555556
},
{
"epoch": 0.19831223628691982,
"grad_norm": 0.62890625,
"learning_rate": 9.84004362809431e-07,
"loss": 0.2462070733308792,
"step": 940,
"token_acc": 0.9301426872770512
},
{
"epoch": 0.19852320675105486,
"grad_norm": 0.81640625,
"learning_rate": 9.839619090028173e-07,
"loss": 0.28827589750289917,
"step": 941,
"token_acc": 0.9256148770245951
},
{
"epoch": 0.19873417721518988,
"grad_norm": 0.7265625,
"learning_rate": 9.83919399851063e-07,
"loss": 0.27797433733940125,
"step": 942,
"token_acc": 0.9241207421766824
},
{
"epoch": 0.1989451476793249,
"grad_norm": 0.84765625,
"learning_rate": 9.838768353590297e-07,
"loss": 0.3198699951171875,
"step": 943,
"token_acc": 0.909556313993174
},
{
"epoch": 0.1991561181434599,
"grad_norm": 0.625,
"learning_rate": 9.838342155315847e-07,
"loss": 0.23603345453739166,
"step": 944,
"token_acc": 0.9322977725674091
},
{
"epoch": 0.19936708860759494,
"grad_norm": 0.63671875,
"learning_rate": 9.837915403736017e-07,
"loss": 0.1939564049243927,
"step": 945,
"token_acc": 0.9405116002379535
},
{
"epoch": 0.19957805907172996,
"grad_norm": 0.6953125,
"learning_rate": 9.837488098899616e-07,
"loss": 0.2682676911354065,
"step": 946,
"token_acc": 0.9287606711804534
},
{
"epoch": 0.19978902953586497,
"grad_norm": 0.94921875,
"learning_rate": 9.837060240855506e-07,
"loss": 0.264107882976532,
"step": 947,
"token_acc": 0.9270292429625581
},
{
"epoch": 0.2,
"grad_norm": 0.78515625,
"learning_rate": 9.836631829652617e-07,
"loss": 0.31936952471733093,
"step": 948,
"token_acc": 0.9171994884910486
},
{
"epoch": 0.20021097046413502,
"grad_norm": 0.76953125,
"learning_rate": 9.83620286533994e-07,
"loss": 0.2610703706741333,
"step": 949,
"token_acc": 0.9247813411078717
},
{
"epoch": 0.20042194092827004,
"grad_norm": 0.76953125,
"learning_rate": 9.835773347966535e-07,
"loss": 0.27383172512054443,
"step": 950,
"token_acc": 0.9266730707652898
},
{
"epoch": 0.20063291139240505,
"grad_norm": 0.7578125,
"learning_rate": 9.835343277581513e-07,
"loss": 0.253266841173172,
"step": 951,
"token_acc": 0.9326241134751773
},
{
"epoch": 0.2008438818565401,
"grad_norm": 0.6875,
"learning_rate": 9.834912654234065e-07,
"loss": 0.24679061770439148,
"step": 952,
"token_acc": 0.9246597024374802
},
{
"epoch": 0.2010548523206751,
"grad_norm": 0.92578125,
"learning_rate": 9.834481477973433e-07,
"loss": 0.25128480792045593,
"step": 953,
"token_acc": 0.9299694189602447
},
{
"epoch": 0.20126582278481012,
"grad_norm": 0.7578125,
"learning_rate": 9.834049748848924e-07,
"loss": 0.26062366366386414,
"step": 954,
"token_acc": 0.9283121597096189
},
{
"epoch": 0.20147679324894516,
"grad_norm": 0.765625,
"learning_rate": 9.833617466909912e-07,
"loss": 0.2557450234889984,
"step": 955,
"token_acc": 0.92808867261422
},
{
"epoch": 0.20168776371308017,
"grad_norm": 0.7265625,
"learning_rate": 9.83318463220583e-07,
"loss": 0.28644537925720215,
"step": 956,
"token_acc": 0.9155513065646909
},
{
"epoch": 0.20189873417721518,
"grad_norm": 0.9375,
"learning_rate": 9.832751244786178e-07,
"loss": 0.27308011054992676,
"step": 957,
"token_acc": 0.9252837326607818
},
{
"epoch": 0.20210970464135022,
"grad_norm": 0.6640625,
"learning_rate": 9.832317304700517e-07,
"loss": 0.2365753948688507,
"step": 958,
"token_acc": 0.9307948860478044
},
{
"epoch": 0.20232067510548524,
"grad_norm": 1.2265625,
"learning_rate": 9.831882811998472e-07,
"loss": 0.25882843136787415,
"step": 959,
"token_acc": 0.9280114041339986
},
{
"epoch": 0.20253164556962025,
"grad_norm": 0.6171875,
"learning_rate": 9.83144776672973e-07,
"loss": 0.24906384944915771,
"step": 960,
"token_acc": 0.9268585131894485
},
{
"epoch": 0.20274261603375526,
"grad_norm": 1.421875,
"learning_rate": 9.831012168944045e-07,
"loss": 0.25317683815956116,
"step": 961,
"token_acc": 0.9272997032640949
},
{
"epoch": 0.2029535864978903,
"grad_norm": 1.40625,
"learning_rate": 9.830576018691227e-07,
"loss": 0.2348695993423462,
"step": 962,
"token_acc": 0.9317293233082706
},
{
"epoch": 0.20316455696202532,
"grad_norm": 0.65625,
"learning_rate": 9.830139316021155e-07,
"loss": 0.22149190306663513,
"step": 963,
"token_acc": 0.9351633986928105
},
{
"epoch": 0.20337552742616033,
"grad_norm": 0.734375,
"learning_rate": 9.829702060983772e-07,
"loss": 0.2570660710334778,
"step": 964,
"token_acc": 0.9295127183573398
},
{
"epoch": 0.20358649789029537,
"grad_norm": 1.0234375,
"learning_rate": 9.829264253629079e-07,
"loss": 0.2847985625267029,
"step": 965,
"token_acc": 0.9199036434808793
},
{
"epoch": 0.20379746835443038,
"grad_norm": 0.72265625,
"learning_rate": 9.828825894007146e-07,
"loss": 0.267423540353775,
"step": 966,
"token_acc": 0.9270031365839749
},
{
"epoch": 0.2040084388185654,
"grad_norm": 6.0625,
"learning_rate": 9.8283869821681e-07,
"loss": 0.2526509165763855,
"step": 967,
"token_acc": 0.9267187106522287
},
{
"epoch": 0.2042194092827004,
"grad_norm": 0.66796875,
"learning_rate": 9.827947518162135e-07,
"loss": 0.22644475102424622,
"step": 968,
"token_acc": 0.9335453100158982
},
{
"epoch": 0.20443037974683545,
"grad_norm": 0.859375,
"learning_rate": 9.827507502039507e-07,
"loss": 0.313146710395813,
"step": 969,
"token_acc": 0.9218163195629908
},
{
"epoch": 0.20464135021097046,
"grad_norm": 0.72265625,
"learning_rate": 9.82706693385054e-07,
"loss": 0.2515157163143158,
"step": 970,
"token_acc": 0.9318849089841457
},
{
"epoch": 0.20485232067510548,
"grad_norm": 0.7890625,
"learning_rate": 9.82662581364561e-07,
"loss": 0.33044931292533875,
"step": 971,
"token_acc": 0.911062906724512
},
{
"epoch": 0.20506329113924052,
"grad_norm": 1.09375,
"learning_rate": 9.826184141475165e-07,
"loss": 0.3272978961467743,
"step": 972,
"token_acc": 0.9140117537890504
},
{
"epoch": 0.20527426160337553,
"grad_norm": 0.59375,
"learning_rate": 9.825741917389717e-07,
"loss": 0.21767356991767883,
"step": 973,
"token_acc": 0.9358974358974359
},
{
"epoch": 0.20548523206751054,
"grad_norm": 0.734375,
"learning_rate": 9.825299141439835e-07,
"loss": 0.28333914279937744,
"step": 974,
"token_acc": 0.9232728430436167
},
{
"epoch": 0.20569620253164558,
"grad_norm": 0.66796875,
"learning_rate": 9.824855813676157e-07,
"loss": 0.23762467503547668,
"step": 975,
"token_acc": 0.9316065192083819
},
{
"epoch": 0.2059071729957806,
"grad_norm": 0.80859375,
"learning_rate": 9.824411934149377e-07,
"loss": 0.2822648882865906,
"step": 976,
"token_acc": 0.9237147595356551
},
{
"epoch": 0.2061181434599156,
"grad_norm": 0.63671875,
"learning_rate": 9.823967502910259e-07,
"loss": 0.2508828043937683,
"step": 977,
"token_acc": 0.9297841726618705
},
{
"epoch": 0.20632911392405062,
"grad_norm": 0.625,
"learning_rate": 9.82352252000963e-07,
"loss": 0.2554951608181,
"step": 978,
"token_acc": 0.9255730872283418
},
{
"epoch": 0.20654008438818566,
"grad_norm": 0.74609375,
"learning_rate": 9.823076985498373e-07,
"loss": 0.2603085935115814,
"step": 979,
"token_acc": 0.927246790299572
},
{
"epoch": 0.20675105485232068,
"grad_norm": 0.75390625,
"learning_rate": 9.82263089942744e-07,
"loss": 0.27707576751708984,
"step": 980,
"token_acc": 0.9242250287026407
},
{
"epoch": 0.2069620253164557,
"grad_norm": 0.74609375,
"learning_rate": 9.822184261847847e-07,
"loss": 0.23693615198135376,
"step": 981,
"token_acc": 0.9334923948702655
},
{
"epoch": 0.20717299578059073,
"grad_norm": 0.69921875,
"learning_rate": 9.821737072810668e-07,
"loss": 0.2479907125234604,
"step": 982,
"token_acc": 0.9293939393939394
},
{
"epoch": 0.20738396624472574,
"grad_norm": 0.72265625,
"learning_rate": 9.821289332367043e-07,
"loss": 0.25571757555007935,
"step": 983,
"token_acc": 0.9304549405969285
},
{
"epoch": 0.20759493670886076,
"grad_norm": 0.7421875,
"learning_rate": 9.820841040568177e-07,
"loss": 0.2608758807182312,
"step": 984,
"token_acc": 0.9285266457680251
},
{
"epoch": 0.20780590717299577,
"grad_norm": 0.7578125,
"learning_rate": 9.820392197465335e-07,
"loss": 0.28490036725997925,
"step": 985,
"token_acc": 0.920461445051609
},
{
"epoch": 0.2080168776371308,
"grad_norm": 0.70703125,
"learning_rate": 9.819942803109844e-07,
"loss": 0.2503746449947357,
"step": 986,
"token_acc": 0.9277822689302075
},
{
"epoch": 0.20822784810126582,
"grad_norm": 0.6953125,
"learning_rate": 9.8194928575531e-07,
"loss": 0.2538071870803833,
"step": 987,
"token_acc": 0.9264251614714968
},
{
"epoch": 0.20843881856540084,
"grad_norm": 0.6875,
"learning_rate": 9.819042360846554e-07,
"loss": 0.2641909718513489,
"step": 988,
"token_acc": 0.9284467713787086
},
{
"epoch": 0.20864978902953588,
"grad_norm": 0.75,
"learning_rate": 9.818591313041727e-07,
"loss": 0.2759447395801544,
"step": 989,
"token_acc": 0.9222654081066074
},
{
"epoch": 0.2088607594936709,
"grad_norm": 0.8125,
"learning_rate": 9.818139714190198e-07,
"loss": 0.23161228001117706,
"step": 990,
"token_acc": 0.9333521604066648
},
{
"epoch": 0.2090717299578059,
"grad_norm": 0.765625,
"learning_rate": 9.817687564343615e-07,
"loss": 0.2939218580722809,
"step": 991,
"token_acc": 0.9156313204276221
},
{
"epoch": 0.20928270042194091,
"grad_norm": 0.7265625,
"learning_rate": 9.817234863553681e-07,
"loss": 0.259197473526001,
"step": 992,
"token_acc": 0.9242243436754176
},
{
"epoch": 0.20949367088607596,
"grad_norm": 0.76171875,
"learning_rate": 9.816781611872167e-07,
"loss": 0.27298709750175476,
"step": 993,
"token_acc": 0.9176136363636364
},
{
"epoch": 0.20970464135021097,
"grad_norm": 0.6328125,
"learning_rate": 9.816327809350907e-07,
"loss": 0.2868914008140564,
"step": 994,
"token_acc": 0.9294675216057987
},
{
"epoch": 0.20991561181434598,
"grad_norm": 0.6171875,
"learning_rate": 9.815873456041797e-07,
"loss": 0.26026803255081177,
"step": 995,
"token_acc": 0.9263676432460461
},
{
"epoch": 0.21012658227848102,
"grad_norm": 0.8359375,
"learning_rate": 9.815418551996795e-07,
"loss": 0.2792215049266815,
"step": 996,
"token_acc": 0.9241547365214743
},
{
"epoch": 0.21033755274261604,
"grad_norm": 0.6953125,
"learning_rate": 9.814963097267925e-07,
"loss": 0.23070243000984192,
"step": 997,
"token_acc": 0.9354066985645934
},
{
"epoch": 0.21054852320675105,
"grad_norm": 0.640625,
"learning_rate": 9.814507091907271e-07,
"loss": 0.2509482502937317,
"step": 998,
"token_acc": 0.9299403078856425
},
{
"epoch": 0.2107594936708861,
"grad_norm": 0.73828125,
"learning_rate": 9.814050535966981e-07,
"loss": 0.24006497859954834,
"step": 999,
"token_acc": 0.9315068493150684
},
{
"epoch": 0.2109704641350211,
"grad_norm": 0.83984375,
"learning_rate": 9.813593429499268e-07,
"loss": 0.28949546813964844,
"step": 1000,
"token_acc": 0.9210992907801419
},
{
"epoch": 0.2109704641350211,
"eval_loss": 0.4336377680301666,
"eval_runtime": 245.5659,
"eval_samples_per_second": 137.254,
"eval_steps_per_second": 2.146,
"eval_token_acc": 0.8990386341200753,
"step": 1000
}
],
"logging_steps": 1,
"max_steps": 9480,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"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": 1.5165183244464292e+18,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}