the-beatles / trainer_state.json
AlekseyKorshuk's picture
huggingartists
7612f16
{
"best_metric": 2.4675137996673584,
"best_model_checkpoint": "output/the-beatles/checkpoint-345",
"epoch": 3.0,
"global_step": 345,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 0.00013679786386078908,
"loss": 3.2135,
"step": 5
},
{
"epoch": 0.07,
"learning_rate": 0.00013559617012171195,
"loss": 3.0835,
"step": 10
},
{
"epoch": 0.1,
"learning_rate": 0.00013360900754314024,
"loss": 3.0207,
"step": 15
},
{
"epoch": 0.14,
"learning_rate": 0.00013085967378942767,
"loss": 2.9809,
"step": 20
},
{
"epoch": 0.17,
"learning_rate": 0.0001273804022850966,
"loss": 3.0379,
"step": 25
},
{
"epoch": 0.21,
"learning_rate": 0.00012321198430740717,
"loss": 2.7644,
"step": 30
},
{
"epoch": 0.24,
"learning_rate": 0.00011840329074592678,
"loss": 2.8189,
"step": 35
},
{
"epoch": 0.28,
"learning_rate": 0.00011301069913603339,
"loss": 2.6556,
"step": 40
},
{
"epoch": 0.31,
"learning_rate": 0.00010709743268385944,
"loss": 2.9363,
"step": 45
},
{
"epoch": 0.34,
"learning_rate": 0.00010073281903200561,
"loss": 2.5794,
"step": 50
},
{
"epoch": 0.38,
"learning_rate": 9.399147745631814e-05,
"loss": 2.4779,
"step": 55
},
{
"epoch": 0.41,
"learning_rate": 8.695244402310457e-05,
"loss": 2.9374,
"step": 60
},
{
"epoch": 0.45,
"learning_rate": 7.969824496351968e-05,
"loss": 2.6178,
"step": 65
},
{
"epoch": 0.48,
"learning_rate": 7.231392912895965e-05,
"loss": 2.7956,
"step": 70
},
{
"epoch": 0.52,
"learning_rate": 6.488607087104036e-05,
"loss": 2.7369,
"step": 75
},
{
"epoch": 0.55,
"learning_rate": 5.750175503648034e-05,
"loss": 2.8983,
"step": 80
},
{
"epoch": 0.59,
"learning_rate": 5.0247555976895465e-05,
"loss": 2.6613,
"step": 85
},
{
"epoch": 0.62,
"learning_rate": 4.320852254368187e-05,
"loss": 2.8101,
"step": 90
},
{
"epoch": 0.66,
"learning_rate": 3.64671809679944e-05,
"loss": 2.7433,
"step": 95
},
{
"epoch": 0.69,
"learning_rate": 3.0102567316140575e-05,
"loss": 2.6317,
"step": 100
},
{
"epoch": 0.72,
"learning_rate": 2.418930086396663e-05,
"loss": 2.8128,
"step": 105
},
{
"epoch": 0.76,
"learning_rate": 1.879670925407324e-05,
"loss": 2.4389,
"step": 110
},
{
"epoch": 0.79,
"learning_rate": 1.3988015692592839e-05,
"loss": 2.6243,
"step": 115
},
{
"epoch": 0.83,
"learning_rate": 9.819597714903392e-06,
"loss": 2.6688,
"step": 120
},
{
"epoch": 0.86,
"learning_rate": 6.340326210572357e-06,
"loss": 2.5304,
"step": 125
},
{
"epoch": 0.9,
"learning_rate": 3.590992456859759e-06,
"loss": 2.6001,
"step": 130
},
{
"epoch": 0.93,
"learning_rate": 1.6038298782880553e-06,
"loss": 2.5575,
"step": 135
},
{
"epoch": 0.97,
"learning_rate": 4.0213613921093164e-07,
"loss": 2.6308,
"step": 140
},
{
"epoch": 1.0,
"learning_rate": 0.0,
"loss": 2.8741,
"step": 145
},
{
"epoch": 1.2,
"learning_rate": 1.3101434185878598e-05,
"loss": 2.452,
"step": 150
},
{
"epoch": 1.24,
"learning_rate": 1.8592752158891168e-05,
"loss": 2.5133,
"step": 155
},
{
"epoch": 1.28,
"learning_rate": 2.4872714303239897e-05,
"loss": 2.3717,
"step": 160
},
{
"epoch": 1.32,
"learning_rate": 3.184228186444085e-05,
"loss": 2.5966,
"step": 165
},
{
"epoch": 1.36,
"learning_rate": 3.939154059863605e-05,
"loss": 2.5,
"step": 170
},
{
"epoch": 1.4,
"learning_rate": 4.74014341858786e-05,
"loss": 2.4988,
"step": 175
},
{
"epoch": 1.44,
"learning_rate": 5.574564181941929e-05,
"loss": 2.6535,
"step": 180
},
{
"epoch": 1.48,
"learning_rate": 6.429257036028911e-05,
"loss": 2.5217,
"step": 185
},
{
"epoch": 1.52,
"learning_rate": 7.290742963971086e-05,
"loss": 2.5988,
"step": 190
},
{
"epoch": 1.56,
"learning_rate": 8.145435818058069e-05,
"loss": 2.3431,
"step": 195
},
{
"epoch": 1.6,
"learning_rate": 8.979856581412138e-05,
"loss": 2.5257,
"step": 200
},
{
"epoch": 1.64,
"learning_rate": 9.780845940136392e-05,
"loss": 2.5586,
"step": 205
},
{
"epoch": 1.68,
"learning_rate": 0.00010535771813555913,
"loss": 2.4987,
"step": 210
},
{
"epoch": 1.72,
"learning_rate": 0.00011232728569676008,
"loss": 2.1873,
"step": 215
},
{
"epoch": 1.76,
"learning_rate": 0.00011860724784110881,
"loss": 2.4757,
"step": 220
},
{
"epoch": 1.8,
"learning_rate": 0.0001240985658141214,
"loss": 2.2676,
"step": 225
},
{
"epoch": 1.84,
"learning_rate": 0.00012871463825100903,
"loss": 2.5652,
"step": 230
},
{
"epoch": 1.88,
"learning_rate": 0.00013238266693193403,
"loss": 2.3751,
"step": 235
},
{
"epoch": 1.92,
"learning_rate": 0.0001350448048534241,
"loss": 2.5399,
"step": 240
},
{
"epoch": 1.96,
"learning_rate": 0.00013665906851017317,
"loss": 2.0634,
"step": 245
},
{
"epoch": 2.0,
"learning_rate": 0.0001372,
"loss": 2.3758,
"step": 250
},
{
"epoch": 2.0,
"eval_loss": 2.5396728515625,
"eval_runtime": 6.8424,
"eval_samples_per_second": 22.653,
"eval_steps_per_second": 2.923,
"step": 250
},
{
"epoch": 2.22,
"learning_rate": 0.0001218137945423232,
"loss": 2.2031,
"step": 255
},
{
"epoch": 2.26,
"learning_rate": 0.00011542314562479972,
"loss": 2.0681,
"step": 260
},
{
"epoch": 2.3,
"learning_rate": 0.0001081602700970799,
"loss": 2.2497,
"step": 265
},
{
"epoch": 2.35,
"learning_rate": 0.00010016046158835702,
"loss": 2.1648,
"step": 270
},
{
"epoch": 2.39,
"learning_rate": 9.157274139492967e-05,
"loss": 2.2327,
"step": 275
},
{
"epoch": 2.43,
"learning_rate": 8.255708249541068e-05,
"loss": 2.2978,
"step": 280
},
{
"epoch": 2.48,
"learning_rate": 7.328142955681647e-05,
"loss": 2.2115,
"step": 285
},
{
"epoch": 2.52,
"learning_rate": 6.391857044318358e-05,
"loss": 2.2139,
"step": 290
},
{
"epoch": 2.57,
"learning_rate": 5.464291750458936e-05,
"loss": 2.3337,
"step": 295
},
{
"epoch": 2.61,
"learning_rate": 4.5627258605070364e-05,
"loss": 2.0981,
"step": 300
},
{
"epoch": 2.65,
"learning_rate": 3.703953841164307e-05,
"loss": 2.1107,
"step": 305
},
{
"epoch": 2.7,
"learning_rate": 2.9039729902920186e-05,
"loss": 2.1893,
"step": 310
},
{
"epoch": 2.74,
"learning_rate": 2.177685437520028e-05,
"loss": 2.353,
"step": 315
},
{
"epoch": 2.78,
"learning_rate": 1.5386205457676823e-05,
"loss": 2.1984,
"step": 320
},
{
"epoch": 2.83,
"learning_rate": 9.986828848110892e-06,
"loss": 2.209,
"step": 325
},
{
"epoch": 2.87,
"learning_rate": 5.679304716725967e-06,
"loss": 2.2126,
"step": 330
},
{
"epoch": 2.91,
"learning_rate": 2.5438740879409566e-06,
"loss": 2.1803,
"step": 335
},
{
"epoch": 2.96,
"learning_rate": 6.389441019077178e-07,
"loss": 2.0247,
"step": 340
},
{
"epoch": 3.0,
"learning_rate": 0.0,
"loss": 2.3114,
"step": 345
},
{
"epoch": 3.0,
"eval_loss": 2.4675137996673584,
"eval_runtime": 2.9387,
"eval_samples_per_second": 80.649,
"eval_steps_per_second": 10.209,
"step": 345
}
],
"max_steps": 345,
"num_train_epochs": 3,
"total_flos": 359015251968000.0,
"trial_name": null,
"trial_params": null
}