KexuanShi's picture
Upload checkpoint-378
9b2dfdf verified
Raw History Blame Contribute Delete
11.6 kB
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0,
"eval_steps": 500,
"global_step": 378,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 1.2536803647875785,
"epoch": 0.05291005291005291,
"grad_norm": 8.8125,
"learning_rate": 3.2142857142857147e-06,
"loss": 1.1722795486450195,
"mean_token_accuracy": 0.7258133076131343,
"num_tokens": 710093.0,
"step": 10
},
{
"entropy": 0.9760579235851765,
"epoch": 0.10582010582010581,
"grad_norm": 2.65625,
"learning_rate": 6.785714285714287e-06,
"loss": 0.9763177871704102,
"mean_token_accuracy": 0.7567757397890091,
"num_tokens": 1429271.0,
"step": 20
},
{
"entropy": 0.794147289544344,
"epoch": 0.15873015873015872,
"grad_norm": 2.46875,
"learning_rate": 1.0357142857142859e-05,
"loss": 0.8046525001525879,
"mean_token_accuracy": 0.7873034350574016,
"num_tokens": 2149735.0,
"step": 30
},
{
"entropy": 0.7088500954210758,
"epoch": 0.21164021164021163,
"grad_norm": 1.4609375,
"learning_rate": 1.3928571428571429e-05,
"loss": 0.7121071815490723,
"mean_token_accuracy": 0.8070594437420369,
"num_tokens": 2868867.0,
"step": 40
},
{
"entropy": 0.683020292967558,
"epoch": 0.26455026455026454,
"grad_norm": 1.46875,
"learning_rate": 1.7500000000000002e-05,
"loss": 0.68443603515625,
"mean_token_accuracy": 0.8138657048344612,
"num_tokens": 3582476.0,
"step": 50
},
{
"entropy": 0.6564815938472748,
"epoch": 0.31746031746031744,
"grad_norm": 1.4296875,
"learning_rate": 1.999986795783372e-05,
"loss": 0.6578450202941895,
"mean_token_accuracy": 0.8192978233098984,
"num_tokens": 4287689.0,
"step": 60
},
{
"entropy": 0.6446133196353913,
"epoch": 0.37037037037037035,
"grad_norm": 1.5859375,
"learning_rate": 1.9997520638548385e-05,
"loss": 0.651158332824707,
"mean_token_accuracy": 0.8207244381308556,
"num_tokens": 5011191.0,
"step": 70
},
{
"entropy": 0.6360202118754387,
"epoch": 0.42328042328042326,
"grad_norm": 1.3984375,
"learning_rate": 1.9992239841688418e-05,
"loss": 0.6350794792175293,
"mean_token_accuracy": 0.8251068785786628,
"num_tokens": 5714842.0,
"step": 80
},
{
"entropy": 0.6189912840723991,
"epoch": 0.47619047619047616,
"grad_norm": 1.328125,
"learning_rate": 1.9984027116747914e-05,
"loss": 0.622182559967041,
"mean_token_accuracy": 0.8279523886740208,
"num_tokens": 6424614.0,
"step": 90
},
{
"entropy": 0.6159785229712724,
"epoch": 0.5291005291005291,
"grad_norm": 1.3671875,
"learning_rate": 1.997288487350878e-05,
"loss": 0.617555046081543,
"mean_token_accuracy": 0.8284852005541324,
"num_tokens": 7139778.0,
"step": 100
},
{
"entropy": 0.6129939660429955,
"epoch": 0.582010582010582,
"grad_norm": 1.328125,
"learning_rate": 1.9958816381333684e-05,
"loss": 0.6128347873687744,
"mean_token_accuracy": 0.8294584318995476,
"num_tokens": 7852540.0,
"step": 110
},
{
"entropy": 0.6052542962133884,
"epoch": 0.6349206349206349,
"grad_norm": 1.3359375,
"learning_rate": 1.994182576820673e-05,
"loss": 0.6046356201171875,
"mean_token_accuracy": 0.831394524872303,
"num_tokens": 8568056.0,
"step": 120
},
{
"entropy": 0.6013224326074124,
"epoch": 0.6878306878306878,
"grad_norm": 1.3203125,
"learning_rate": 1.9921918019522238e-05,
"loss": 0.6063410758972168,
"mean_token_accuracy": 0.8313738636672496,
"num_tokens": 9276742.0,
"step": 130
},
{
"entropy": 0.5983927462249994,
"epoch": 0.7407407407407407,
"grad_norm": 1.3125,
"learning_rate": 1.9899098976621922e-05,
"loss": 0.5977112770080566,
"mean_token_accuracy": 0.8335464030504227,
"num_tokens": 9990163.0,
"step": 140
},
{
"entropy": 0.5922725383192301,
"epoch": 0.7936507936507936,
"grad_norm": 1.5078125,
"learning_rate": 1.9873375335080915e-05,
"loss": 0.5917873382568359,
"mean_token_accuracy": 0.8347524181008339,
"num_tokens": 10710006.0,
"step": 150
},
{
"entropy": 0.5920293003320694,
"epoch": 0.8465608465608465,
"grad_norm": 1.3046875,
"learning_rate": 1.9844754642743164e-05,
"loss": 0.5962053298950195,
"mean_token_accuracy": 0.8335559628903866,
"num_tokens": 11416822.0,
"step": 160
},
{
"entropy": 0.5966673269867897,
"epoch": 0.8994708994708994,
"grad_norm": 1.3828125,
"learning_rate": 1.981324529750672e-05,
"loss": 0.6019061088562012,
"mean_token_accuracy": 0.83258865326643,
"num_tokens": 12123118.0,
"step": 170
},
{
"entropy": 0.5854565776884556,
"epoch": 0.9523809523809523,
"grad_norm": 1.359375,
"learning_rate": 1.9778856544859647e-05,
"loss": 0.5845203399658203,
"mean_token_accuracy": 0.8360438510775566,
"num_tokens": 12831058.0,
"step": 180
},
{
"entropy": 0.5759805988520383,
"epoch": 1.0052910052910053,
"grad_norm": 1.3046875,
"learning_rate": 1.9741598475167173e-05,
"loss": 0.5696067810058594,
"mean_token_accuracy": 0.839682175219059,
"num_tokens": 13547943.0,
"step": 190
},
{
"entropy": 0.5275494448840619,
"epoch": 1.0582010582010581,
"grad_norm": 1.3984375,
"learning_rate": 1.9701482020711004e-05,
"loss": 0.5299928665161133,
"mean_token_accuracy": 0.8484233327209949,
"num_tokens": 14264772.0,
"step": 200
},
{
"entropy": 0.5196751751005649,
"epoch": 1.1111111111111112,
"grad_norm": 1.2109375,
"learning_rate": 1.9658518952481543e-05,
"loss": 0.5208037853240967,
"mean_token_accuracy": 0.8509362854063511,
"num_tokens": 14985565.0,
"step": 210
},
{
"entropy": 0.5303940393030644,
"epoch": 1.164021164021164,
"grad_norm": 1.3359375,
"learning_rate": 1.9612721876724054e-05,
"loss": 0.5318955421447754,
"mean_token_accuracy": 0.8480072498321534,
"num_tokens": 15701796.0,
"step": 220
},
{
"entropy": 0.5232694145292044,
"epoch": 1.216931216931217,
"grad_norm": 1.359375,
"learning_rate": 1.9564104231239726e-05,
"loss": 0.5231654167175293,
"mean_token_accuracy": 0.8490844376385212,
"num_tokens": 16412522.0,
"step": 230
},
{
"entropy": 0.529466836899519,
"epoch": 1.2698412698412698,
"grad_norm": 1.3046875,
"learning_rate": 1.9512680281442745e-05,
"loss": 0.5309067726135254,
"mean_token_accuracy": 0.8485323786735535,
"num_tokens": 17126523.0,
"step": 240
},
{
"entropy": 0.527790617197752,
"epoch": 1.3227513227513228,
"grad_norm": 1.3125,
"learning_rate": 1.9458465116174534e-05,
"loss": 0.5262090682983398,
"mean_token_accuracy": 0.8493455089628696,
"num_tokens": 17834934.0,
"step": 250
},
{
"entropy": 0.528411227837205,
"epoch": 1.3756613756613756,
"grad_norm": 1.265625,
"learning_rate": 1.940147464327637e-05,
"loss": 0.529714298248291,
"mean_token_accuracy": 0.8480914957821369,
"num_tokens": 18553413.0,
"step": 260
},
{
"entropy": 0.516933124884963,
"epoch": 1.4285714285714286,
"grad_norm": 1.296875,
"learning_rate": 1.9341725584921712e-05,
"loss": 0.5172879219055175,
"mean_token_accuracy": 0.8517736725509166,
"num_tokens": 19270191.0,
"step": 270
},
{
"entropy": 0.5223661050200462,
"epoch": 1.4814814814814814,
"grad_norm": 1.2421875,
"learning_rate": 1.9279235472709554e-05,
"loss": 0.5231949329376221,
"mean_token_accuracy": 0.8496842503547668,
"num_tokens": 19976552.0,
"step": 280
},
{
"entropy": 0.5293067336082459,
"epoch": 1.5343915343915344,
"grad_norm": 1.2421875,
"learning_rate": 1.92140226425203e-05,
"loss": 0.5295753479003906,
"mean_token_accuracy": 0.8479710258543491,
"num_tokens": 20676613.0,
"step": 290
},
{
"entropy": 0.5302139949053526,
"epoch": 1.5873015873015874,
"grad_norm": 1.28125,
"learning_rate": 1.914610622913566e-05,
"loss": 0.530374813079834,
"mean_token_accuracy": 0.8484688423573971,
"num_tokens": 21385978.0,
"step": 300
},
{
"entropy": 0.5183596257120371,
"epoch": 1.6402116402116402,
"grad_norm": 1.1953125,
"learning_rate": 1.9075506160624068e-05,
"loss": 0.519154167175293,
"mean_token_accuracy": 0.8509341597557067,
"num_tokens": 22094336.0,
"step": 310
},
{
"entropy": 0.5201449614018202,
"epoch": 1.693121693121693,
"grad_norm": 1.328125,
"learning_rate": 1.900224315249343e-05,
"loss": 0.5171724796295166,
"mean_token_accuracy": 0.8508495308458806,
"num_tokens": 22798906.0,
"step": 320
},
{
"entropy": 0.5111908499151468,
"epoch": 1.746031746031746,
"grad_norm": 1.2265625,
"learning_rate": 1.8926338701612738e-05,
"loss": 0.511273717880249,
"mean_token_accuracy": 0.8524932019412518,
"num_tokens": 23511239.0,
"step": 330
},
{
"entropy": 0.5265866108238697,
"epoch": 1.798941798941799,
"grad_norm": 1.265625,
"learning_rate": 1.884781507990445e-05,
"loss": 0.5289421558380127,
"mean_token_accuracy": 0.8486531592905522,
"num_tokens": 24231185.0,
"step": 340
},
{
"entropy": 0.5189403794705868,
"epoch": 1.8518518518518519,
"grad_norm": 1.2578125,
"learning_rate": 1.8766695327809463e-05,
"loss": 0.5158460140228271,
"mean_token_accuracy": 0.8517816573381424,
"num_tokens": 24944285.0,
"step": 350
},
{
"entropy": 0.5055529039353133,
"epoch": 1.9047619047619047,
"grad_norm": 1.1953125,
"learning_rate": 1.8683003247526567e-05,
"loss": 0.5064782619476318,
"mean_token_accuracy": 0.8533773042261601,
"num_tokens": 25670441.0,
"step": 360
},
{
"entropy": 0.5245815236121416,
"epoch": 1.9576719576719577,
"grad_norm": 1.2421875,
"learning_rate": 1.8596763396028384e-05,
"loss": 0.5262872219085694,
"mean_token_accuracy": 0.8492341041564941,
"num_tokens": 26379786.0,
"step": 370
}
],
"logging_steps": 10,
"max_steps": 1890,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"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": 4.860945833530491e+17,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}