checkpoints-2-epochs / trainer_state.json
3sara's picture
from 388 checkpoint
d5627f3 verified
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 5.984615384615385,
"eval_steps": 500,
"global_step": 582,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.010256410256410256,
"eval_loss": 0.3835288882255554,
"eval_runtime": 178.3246,
"eval_samples_per_second": 1.155,
"eval_steps_per_second": 0.292,
"step": 1
},
{
"epoch": 0.20512820512820512,
"grad_norm": 6.791983604431152,
"learning_rate": 9.5e-06,
"loss": 0.2452,
"step": 20
},
{
"epoch": 0.41025641025641024,
"grad_norm": 11.432096481323242,
"learning_rate": 1.9500000000000003e-05,
"loss": 0.1853,
"step": 40
},
{
"epoch": 0.6153846153846154,
"grad_norm": 1.4045106172561646,
"learning_rate": 2.95e-05,
"loss": 0.1126,
"step": 60
},
{
"epoch": 0.8205128205128205,
"grad_norm": 11.10899543762207,
"learning_rate": 3.9500000000000005e-05,
"loss": 0.2361,
"step": 80
},
{
"epoch": 1.0205128205128204,
"grad_norm": 2.5366597175598145,
"learning_rate": 4.9500000000000004e-05,
"loss": 0.1393,
"step": 100
},
{
"epoch": 1.2256410256410257,
"grad_norm": 19.193058013916016,
"learning_rate": 3.9893617021276594e-05,
"loss": 0.182,
"step": 120
},
{
"epoch": 1.4307692307692308,
"grad_norm": 2.2622101306915283,
"learning_rate": 2.925531914893617e-05,
"loss": 0.0682,
"step": 140
},
{
"epoch": 1.6358974358974359,
"grad_norm": 1.3008161783218384,
"learning_rate": 1.8617021276595745e-05,
"loss": 0.0844,
"step": 160
},
{
"epoch": 1.8410256410256411,
"grad_norm": 1.2486586570739746,
"learning_rate": 7.97872340425532e-06,
"loss": 0.1358,
"step": 180
},
{
"epoch": 2.0615384615384613,
"grad_norm": 0.1917697638273239,
"learning_rate": 2.4083769633507854e-05,
"loss": 0.0433,
"step": 200
},
{
"epoch": 2.2666666666666666,
"grad_norm": 0.0892116129398346,
"learning_rate": 2.934027777777778e-05,
"loss": 0.0348,
"step": 220
},
{
"epoch": 2.471794871794872,
"grad_norm": 1.184501051902771,
"learning_rate": 2.5868055555555558e-05,
"loss": 0.0639,
"step": 240
},
{
"epoch": 2.676923076923077,
"grad_norm": 0.4964236617088318,
"learning_rate": 2.2395833333333337e-05,
"loss": 0.1196,
"step": 260
},
{
"epoch": 2.882051282051282,
"grad_norm": 8.024321556091309,
"learning_rate": 1.8923611111111112e-05,
"loss": 0.0912,
"step": 280
},
{
"epoch": 3.0923076923076924,
"grad_norm": 0.8915512561798096,
"learning_rate": 1.545138888888889e-05,
"loss": 0.0188,
"step": 300
},
{
"epoch": 3.2974358974358973,
"grad_norm": 0.3651086091995239,
"learning_rate": 1.1979166666666667e-05,
"loss": 0.0696,
"step": 320
},
{
"epoch": 3.5025641025641026,
"grad_norm": 7.531687259674072,
"learning_rate": 8.506944444444445e-06,
"loss": 0.0439,
"step": 340
},
{
"epoch": 3.707692307692308,
"grad_norm": 16.617700576782227,
"learning_rate": 5.034722222222222e-06,
"loss": 0.0434,
"step": 360
},
{
"epoch": 3.9128205128205127,
"grad_norm": 0.07135667651891708,
"learning_rate": 1.5625e-06,
"loss": 0.0137,
"step": 380
},
{
"epoch": 4.123076923076923,
"grad_norm": 4.136926174163818,
"learning_rate": 1.8983402489626556e-05,
"loss": 0.1377,
"step": 400
},
{
"epoch": 4.328205128205128,
"grad_norm": 0.3030303716659546,
"learning_rate": 1.690871369294606e-05,
"loss": 0.0382,
"step": 420
},
{
"epoch": 4.533333333333333,
"grad_norm": 0.07416559755802155,
"learning_rate": 1.4834024896265561e-05,
"loss": 0.0615,
"step": 440
},
{
"epoch": 4.7384615384615385,
"grad_norm": 6.657896995544434,
"learning_rate": 1.2759336099585062e-05,
"loss": 0.0241,
"step": 460
},
{
"epoch": 4.943589743589744,
"grad_norm": 0.006628558039665222,
"learning_rate": 1.0684647302904565e-05,
"loss": 0.0742,
"step": 480
},
{
"epoch": 5.143589743589744,
"grad_norm": 0.06650497764348984,
"learning_rate": 8.609958506224066e-06,
"loss": 0.0863,
"step": 500
},
{
"epoch": 5.143589743589744,
"eval_loss": 0.3170950710773468,
"eval_runtime": 176.7917,
"eval_samples_per_second": 1.165,
"eval_steps_per_second": 0.294,
"step": 500
},
{
"epoch": 5.348717948717948,
"grad_norm": 0.06573180109262466,
"learning_rate": 6.535269709543569e-06,
"loss": 0.0203,
"step": 520
},
{
"epoch": 5.553846153846154,
"grad_norm": 0.2770240604877472,
"learning_rate": 4.460580912863071e-06,
"loss": 0.0249,
"step": 540
},
{
"epoch": 5.758974358974359,
"grad_norm": 0.49672994017601013,
"learning_rate": 2.3858921161825725e-06,
"loss": 0.0353,
"step": 560
},
{
"epoch": 5.964102564102564,
"grad_norm": 0.03965625539422035,
"learning_rate": 3.112033195020747e-07,
"loss": 0.052,
"step": 580
},
{
"epoch": 5.984615384615385,
"step": 582,
"total_flos": 1.401094867087152e+17,
"train_loss": 0.017203848492162128,
"train_runtime": 4704.2352,
"train_samples_per_second": 1.987,
"train_steps_per_second": 0.124
}
],
"logging_steps": 20,
"max_steps": 582,
"num_input_tokens_seen": 0,
"num_train_epochs": 6,
"save_steps": 200,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.401094867087152e+17,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}