sft_LIMA / trainer_state.json
Xinging's picture
Upload trainer_state.json with huggingface_hub
a423b25 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 99,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.030303030303030304,
"grad_norm": 1.5466958284378052,
"learning_rate": 2.0000000000000003e-06,
"loss": 1.8566,
"step": 1
},
{
"epoch": 0.06060606060606061,
"grad_norm": 1.5582852363586426,
"learning_rate": 4.000000000000001e-06,
"loss": 1.8976,
"step": 2
},
{
"epoch": 0.09090909090909091,
"grad_norm": 1.524529218673706,
"learning_rate": 6e-06,
"loss": 1.7259,
"step": 3
},
{
"epoch": 0.12121212121212122,
"grad_norm": 1.6974282264709473,
"learning_rate": 8.000000000000001e-06,
"loss": 1.814,
"step": 4
},
{
"epoch": 0.15151515151515152,
"grad_norm": 1.4745838642120361,
"learning_rate": 1e-05,
"loss": 1.5887,
"step": 5
},
{
"epoch": 0.18181818181818182,
"grad_norm": 1.522696852684021,
"learning_rate": 1.2e-05,
"loss": 1.692,
"step": 6
},
{
"epoch": 0.21212121212121213,
"grad_norm": 1.4861280918121338,
"learning_rate": 1.4e-05,
"loss": 1.6883,
"step": 7
},
{
"epoch": 0.24242424242424243,
"grad_norm": 1.441834807395935,
"learning_rate": 1.6000000000000003e-05,
"loss": 1.6722,
"step": 8
},
{
"epoch": 0.2727272727272727,
"grad_norm": 1.633545160293579,
"learning_rate": 1.8e-05,
"loss": 1.6187,
"step": 9
},
{
"epoch": 0.30303030303030304,
"grad_norm": 1.495406150817871,
"learning_rate": 2e-05,
"loss": 1.5615,
"step": 10
},
{
"epoch": 0.3333333333333333,
"grad_norm": 1.5129448175430298,
"learning_rate": 1.9993770622619784e-05,
"loss": 1.6768,
"step": 11
},
{
"epoch": 0.36363636363636365,
"grad_norm": 1.461745023727417,
"learning_rate": 1.9975090251507637e-05,
"loss": 1.7377,
"step": 12
},
{
"epoch": 0.3939393939393939,
"grad_norm": 1.2962136268615723,
"learning_rate": 1.9943982160079823e-05,
"loss": 1.7039,
"step": 13
},
{
"epoch": 0.42424242424242425,
"grad_norm": 1.5274053812026978,
"learning_rate": 1.9900485105144544e-05,
"loss": 1.8094,
"step": 14
},
{
"epoch": 0.45454545454545453,
"grad_norm": 1.8526736497879028,
"learning_rate": 1.9844653278615836e-05,
"loss": 1.7952,
"step": 15
},
{
"epoch": 0.48484848484848486,
"grad_norm": 1.5075675249099731,
"learning_rate": 1.9776556239997146e-05,
"loss": 1.7614,
"step": 16
},
{
"epoch": 0.5151515151515151,
"grad_norm": 1.3172680139541626,
"learning_rate": 1.9696278829718882e-05,
"loss": 1.8214,
"step": 17
},
{
"epoch": 0.5454545454545454,
"grad_norm": 1.2293428182601929,
"learning_rate": 1.9603921063437795e-05,
"loss": 1.688,
"step": 18
},
{
"epoch": 0.5757575757575758,
"grad_norm": 1.6465842723846436,
"learning_rate": 1.949959800742991e-05,
"loss": 1.7641,
"step": 19
},
{
"epoch": 0.6060606060606061,
"grad_norm": 1.2794415950775146,
"learning_rate": 1.9383439635232296e-05,
"loss": 1.7237,
"step": 20
},
{
"epoch": 0.6363636363636364,
"grad_norm": 1.5035333633422852,
"learning_rate": 1.9255590665712214e-05,
"loss": 1.7502,
"step": 21
},
{
"epoch": 0.6666666666666666,
"grad_norm": 1.6453092098236084,
"learning_rate": 1.911621038276542e-05,
"loss": 1.4906,
"step": 22
},
{
"epoch": 0.696969696969697,
"grad_norm": 1.51984703540802,
"learning_rate": 1.8965472436868288e-05,
"loss": 1.5359,
"step": 23
},
{
"epoch": 0.7272727272727273,
"grad_norm": 1.3503923416137695,
"learning_rate": 1.8803564628730916e-05,
"loss": 1.5969,
"step": 24
},
{
"epoch": 0.7575757575757576,
"grad_norm": 1.3228968381881714,
"learning_rate": 1.8630688675320844e-05,
"loss": 1.7511,
"step": 25
},
{
"epoch": 0.7878787878787878,
"grad_norm": 1.2232959270477295,
"learning_rate": 1.8447059958548822e-05,
"loss": 1.6303,
"step": 26
},
{
"epoch": 0.8181818181818182,
"grad_norm": 1.2176752090454102,
"learning_rate": 1.8252907256929777e-05,
"loss": 1.6744,
"step": 27
},
{
"epoch": 0.8484848484848485,
"grad_norm": 1.6348463296890259,
"learning_rate": 1.804847246055326e-05,
"loss": 1.7386,
"step": 28
},
{
"epoch": 0.8787878787878788,
"grad_norm": 1.3479255437850952,
"learning_rate": 1.7834010269718526e-05,
"loss": 1.8378,
"step": 29
},
{
"epoch": 0.9090909090909091,
"grad_norm": 1.4100639820098877,
"learning_rate": 1.7609787877609678e-05,
"loss": 1.6901,
"step": 30
},
{
"epoch": 0.9393939393939394,
"grad_norm": 1.289868950843811,
"learning_rate": 1.7376084637406222e-05,
"loss": 1.7371,
"step": 31
},
{
"epoch": 0.9696969696969697,
"grad_norm": 1.2278428077697754,
"learning_rate": 1.7133191714243805e-05,
"loss": 1.5087,
"step": 32
},
{
"epoch": 1.0,
"grad_norm": 1.5690200328826904,
"learning_rate": 1.6881411722458688e-05,
"loss": 1.4644,
"step": 33
},
{
"epoch": 1.0303030303030303,
"grad_norm": 1.7996729612350464,
"learning_rate": 1.6621058348568008e-05,
"loss": 1.2771,
"step": 34
},
{
"epoch": 1.0606060606060606,
"grad_norm": 1.6773903369903564,
"learning_rate": 1.6352455960455385e-05,
"loss": 1.324,
"step": 35
},
{
"epoch": 1.0909090909090908,
"grad_norm": 1.6019961833953857,
"learning_rate": 1.607593920324899e-05,
"loss": 1.2817,
"step": 36
},
{
"epoch": 1.121212121212121,
"grad_norm": 1.3165000677108765,
"learning_rate": 1.5791852582395334e-05,
"loss": 1.1927,
"step": 37
},
{
"epoch": 1.1515151515151516,
"grad_norm": 1.455589771270752,
"learning_rate": 1.5500550034448415e-05,
"loss": 1.2133,
"step": 38
},
{
"epoch": 1.1818181818181819,
"grad_norm": 1.8257672786712646,
"learning_rate": 1.5202394486108823e-05,
"loss": 1.1531,
"step": 39
},
{
"epoch": 1.2121212121212122,
"grad_norm": 1.873024821281433,
"learning_rate": 1.4897757402062285e-05,
"loss": 1.2731,
"step": 40
},
{
"epoch": 1.2424242424242424,
"grad_norm": 1.7878319025039673,
"learning_rate": 1.4587018322180906e-05,
"loss": 1.2247,
"step": 41
},
{
"epoch": 1.2727272727272727,
"grad_norm": 1.2891494035720825,
"learning_rate": 1.4270564388663761e-05,
"loss": 1.0763,
"step": 42
},
{
"epoch": 1.303030303030303,
"grad_norm": 1.4569677114486694,
"learning_rate": 1.3948789863705914e-05,
"loss": 1.1027,
"step": 43
},
{
"epoch": 1.3333333333333333,
"grad_norm": 1.5054666996002197,
"learning_rate": 1.3622095638296827e-05,
"loss": 1.1478,
"step": 44
},
{
"epoch": 1.3636363636363638,
"grad_norm": 1.4277585744857788,
"learning_rate": 1.32908887327601e-05,
"loss": 1.1792,
"step": 45
},
{
"epoch": 1.393939393939394,
"grad_norm": 1.3641875982284546,
"learning_rate": 1.2955581789656844e-05,
"loss": 1.0412,
"step": 46
},
{
"epoch": 1.4242424242424243,
"grad_norm": 1.706855297088623,
"learning_rate": 1.2616592559684408e-05,
"loss": 1.1198,
"step": 47
},
{
"epoch": 1.4545454545454546,
"grad_norm": 1.8636873960494995,
"learning_rate": 1.2274343381211067e-05,
"loss": 1.0971,
"step": 48
},
{
"epoch": 1.4848484848484849,
"grad_norm": 1.6042084693908691,
"learning_rate": 1.192926065409497e-05,
"loss": 1.0331,
"step": 49
},
{
"epoch": 1.5151515151515151,
"grad_norm": 1.6517517566680908,
"learning_rate": 1.1581774308443042e-05,
"loss": 1.1258,
"step": 50
},
{
"epoch": 1.5454545454545454,
"grad_norm": 1.7735017538070679,
"learning_rate": 1.1232317268971586e-05,
"loss": 1.211,
"step": 51
},
{
"epoch": 1.5757575757575757,
"grad_norm": 1.4876844882965088,
"learning_rate": 1.088132491563602e-05,
"loss": 1.0778,
"step": 52
},
{
"epoch": 1.606060606060606,
"grad_norm": 1.4725524187088013,
"learning_rate": 1.0529234541201631e-05,
"loss": 1.1482,
"step": 53
},
{
"epoch": 1.6363636363636362,
"grad_norm": 1.6460925340652466,
"learning_rate": 1.0176484806431288e-05,
"loss": 1.0857,
"step": 54
},
{
"epoch": 1.6666666666666665,
"grad_norm": 1.6207877397537231,
"learning_rate": 9.823515193568715e-06,
"loss": 1.1458,
"step": 55
},
{
"epoch": 1.696969696969697,
"grad_norm": 1.5556446313858032,
"learning_rate": 9.470765458798369e-06,
"loss": 0.9578,
"step": 56
},
{
"epoch": 1.7272727272727273,
"grad_norm": 1.459346890449524,
"learning_rate": 9.118675084363986e-06,
"loss": 1.1004,
"step": 57
},
{
"epoch": 1.7575757575757576,
"grad_norm": 1.6329230070114136,
"learning_rate": 8.767682731028415e-06,
"loss": 1.0838,
"step": 58
},
{
"epoch": 1.7878787878787878,
"grad_norm": 1.4568471908569336,
"learning_rate": 8.418225691556962e-06,
"loss": 0.9716,
"step": 59
},
{
"epoch": 1.8181818181818183,
"grad_norm": 1.6114221811294556,
"learning_rate": 8.070739345905032e-06,
"loss": 1.2118,
"step": 60
},
{
"epoch": 1.8484848484848486,
"grad_norm": 1.4424853324890137,
"learning_rate": 7.725656618788938e-06,
"loss": 1.0933,
"step": 61
},
{
"epoch": 1.878787878787879,
"grad_norm": 1.3055541515350342,
"learning_rate": 7.383407440315595e-06,
"loss": 1.054,
"step": 62
},
{
"epoch": 1.9090909090909092,
"grad_norm": 1.608759880065918,
"learning_rate": 7.044418210343161e-06,
"loss": 1.2112,
"step": 63
},
{
"epoch": 1.9393939393939394,
"grad_norm": 1.381705403327942,
"learning_rate": 6.7091112672399e-06,
"loss": 1.06,
"step": 64
},
{
"epoch": 1.9696969696969697,
"grad_norm": 1.5174615383148193,
"learning_rate": 6.3779043617031775e-06,
"loss": 1.1465,
"step": 65
},
{
"epoch": 2.0,
"grad_norm": 1.7489651441574097,
"learning_rate": 6.051210136294089e-06,
"loss": 0.8968,
"step": 66
},
{
"epoch": 2.0303030303030303,
"grad_norm": 1.7633519172668457,
"learning_rate": 5.729435611336239e-06,
"loss": 0.8801,
"step": 67
},
{
"epoch": 2.0606060606060606,
"grad_norm": 2.278275728225708,
"learning_rate": 5.412981677819094e-06,
"loss": 0.8784,
"step": 68
},
{
"epoch": 2.090909090909091,
"grad_norm": 1.8437879085540771,
"learning_rate": 5.1022425979377174e-06,
"loss": 0.7264,
"step": 69
},
{
"epoch": 2.121212121212121,
"grad_norm": 1.5237785577774048,
"learning_rate": 4.797605513891179e-06,
"loss": 0.758,
"step": 70
},
{
"epoch": 2.1515151515151514,
"grad_norm": 1.7430484294891357,
"learning_rate": 4.4994499655515865e-06,
"loss": 0.8087,
"step": 71
},
{
"epoch": 2.1818181818181817,
"grad_norm": 1.7282936573028564,
"learning_rate": 4.208147417604665e-06,
"loss": 0.7311,
"step": 72
},
{
"epoch": 2.212121212121212,
"grad_norm": 1.907755970954895,
"learning_rate": 3.924060796751012e-06,
"loss": 0.7755,
"step": 73
},
{
"epoch": 2.242424242424242,
"grad_norm": 2.1145212650299072,
"learning_rate": 3.647544039544615e-06,
"loss": 0.7668,
"step": 74
},
{
"epoch": 2.2727272727272725,
"grad_norm": 2.2172389030456543,
"learning_rate": 3.378941651431996e-06,
"loss": 0.7701,
"step": 75
},
{
"epoch": 2.303030303030303,
"grad_norm": 2.0913803577423096,
"learning_rate": 3.1185882775413123e-06,
"loss": 0.889,
"step": 76
},
{
"epoch": 2.3333333333333335,
"grad_norm": 2.1478118896484375,
"learning_rate": 2.8668082857562006e-06,
"loss": 0.7411,
"step": 77
},
{
"epoch": 2.3636363636363638,
"grad_norm": 2.0034427642822266,
"learning_rate": 2.6239153625937786e-06,
"loss": 0.7564,
"step": 78
},
{
"epoch": 2.393939393939394,
"grad_norm": 1.6710702180862427,
"learning_rate": 2.390212122390323e-06,
"loss": 0.7695,
"step": 79
},
{
"epoch": 2.4242424242424243,
"grad_norm": 1.7885438203811646,
"learning_rate": 2.165989730281475e-06,
"loss": 0.8017,
"step": 80
},
{
"epoch": 2.4545454545454546,
"grad_norm": 1.3610081672668457,
"learning_rate": 1.9515275394467446e-06,
"loss": 0.7561,
"step": 81
},
{
"epoch": 2.484848484848485,
"grad_norm": 1.645119547843933,
"learning_rate": 1.7470927430702277e-06,
"loss": 0.6155,
"step": 82
},
{
"epoch": 2.515151515151515,
"grad_norm": 1.7851331233978271,
"learning_rate": 1.5529400414511809e-06,
"loss": 0.6894,
"step": 83
},
{
"epoch": 2.5454545454545454,
"grad_norm": 1.4505351781845093,
"learning_rate": 1.369311324679159e-06,
"loss": 0.8904,
"step": 84
},
{
"epoch": 2.5757575757575757,
"grad_norm": 1.663177728652954,
"learning_rate": 1.196435371269089e-06,
"loss": 0.8289,
"step": 85
},
{
"epoch": 2.606060606060606,
"grad_norm": 1.7335422039031982,
"learning_rate": 1.0345275631317165e-06,
"loss": 0.7425,
"step": 86
},
{
"epoch": 2.6363636363636362,
"grad_norm": 1.3842012882232666,
"learning_rate": 8.837896172345827e-07,
"loss": 0.7578,
"step": 87
},
{
"epoch": 2.6666666666666665,
"grad_norm": 1.770256757736206,
"learning_rate": 7.4440933428779e-07,
"loss": 0.6974,
"step": 88
},
{
"epoch": 2.6969696969696972,
"grad_norm": 1.5890138149261475,
"learning_rate": 6.165603647677054e-07,
"loss": 0.7797,
"step": 89
},
{
"epoch": 2.7272727272727275,
"grad_norm": 1.6013965606689453,
"learning_rate": 5.004019925700921e-07,
"loss": 0.7053,
"step": 90
},
{
"epoch": 2.757575757575758,
"grad_norm": 1.4727004766464233,
"learning_rate": 3.960789365622075e-07,
"loss": 0.7264,
"step": 91
},
{
"epoch": 2.787878787878788,
"grad_norm": 1.6595274209976196,
"learning_rate": 3.0372117028111825e-07,
"loss": 0.7089,
"step": 92
},
{
"epoch": 2.8181818181818183,
"grad_norm": 1.8651310205459595,
"learning_rate": 2.2344376000285606e-07,
"loss": 0.6721,
"step": 93
},
{
"epoch": 2.8484848484848486,
"grad_norm": 1.7887459993362427,
"learning_rate": 1.553467213841664e-07,
"loss": 0.788,
"step": 94
},
{
"epoch": 2.878787878787879,
"grad_norm": 1.454445719718933,
"learning_rate": 9.951489485545696e-08,
"loss": 0.6617,
"step": 95
},
{
"epoch": 2.909090909090909,
"grad_norm": 1.662096619606018,
"learning_rate": 5.6017839920180506e-08,
"loss": 0.7603,
"step": 96
},
{
"epoch": 2.9393939393939394,
"grad_norm": 1.6245455741882324,
"learning_rate": 2.4909748492362162e-08,
"loss": 0.7336,
"step": 97
},
{
"epoch": 2.9696969696969697,
"grad_norm": 1.5725836753845215,
"learning_rate": 6.229377380218005e-09,
"loss": 0.7566,
"step": 98
},
{
"epoch": 3.0,
"grad_norm": 1.6165506839752197,
"learning_rate": 0.0,
"loss": 0.6936,
"step": 99
},
{
"epoch": 3.0,
"step": 99,
"total_flos": 2.3659576070189875e+17,
"train_loss": 1.195338128191052,
"train_runtime": 1048.1424,
"train_samples_per_second": 2.945,
"train_steps_per_second": 0.094
}
],
"logging_steps": 1,
"max_steps": 99,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 0,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 2.3659576070189875e+17,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}