Files
llama3.2-1b-gsm8k-full/checkpoint-500/trainer_state.json

384 lines
9.1 KiB
JSON
Raw Normal View History

{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9191176470588235,
"eval_steps": 500,
"global_step": 500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01838235294117647,
"grad_norm": 24.461347652326168,
"learning_rate": 6.097560975609757e-07,
"loss": 1.2036,
"step": 10
},
{
"epoch": 0.03676470588235294,
"grad_norm": 17.308880580961944,
"learning_rate": 1.2195121951219514e-06,
"loss": 1.2064,
"step": 20
},
{
"epoch": 0.05514705882352941,
"grad_norm": 29.601720546719697,
"learning_rate": 1.8292682926829268e-06,
"loss": 1.2741,
"step": 30
},
{
"epoch": 0.07352941176470588,
"grad_norm": 14.115818236245202,
"learning_rate": 2.4390243902439027e-06,
"loss": 0.7596,
"step": 40
},
{
"epoch": 0.09191176470588236,
"grad_norm": 21.73939509164672,
"learning_rate": 3.0487804878048782e-06,
"loss": 0.6171,
"step": 50
},
{
"epoch": 0.11029411764705882,
"grad_norm": 16.96116387460838,
"learning_rate": 3.6585365853658537e-06,
"loss": 0.794,
"step": 60
},
{
"epoch": 0.12867647058823528,
"grad_norm": 15.372652400517632,
"learning_rate": 4.268292682926829e-06,
"loss": 0.6233,
"step": 70
},
{
"epoch": 0.14705882352941177,
"grad_norm": 15.062285237489078,
"learning_rate": 4.8780487804878055e-06,
"loss": 0.5441,
"step": 80
},
{
"epoch": 0.16544117647058823,
"grad_norm": 18.632304280702545,
"learning_rate": 5.487804878048781e-06,
"loss": 0.5784,
"step": 90
},
{
"epoch": 0.18382352941176472,
"grad_norm": 13.945519044585177,
"learning_rate": 6.0975609756097564e-06,
"loss": 0.4844,
"step": 100
},
{
"epoch": 0.20220588235294118,
"grad_norm": 6.9751627817275,
"learning_rate": 6.707317073170733e-06,
"loss": 0.6535,
"step": 110
},
{
"epoch": 0.22058823529411764,
"grad_norm": 25.188175477401735,
"learning_rate": 7.317073170731707e-06,
"loss": 0.6897,
"step": 120
},
{
"epoch": 0.23897058823529413,
"grad_norm": 14.572936912172068,
"learning_rate": 7.926829268292685e-06,
"loss": 0.4217,
"step": 130
},
{
"epoch": 0.25735294117647056,
"grad_norm": 14.934836192557746,
"learning_rate": 8.536585365853658e-06,
"loss": 0.5868,
"step": 140
},
{
"epoch": 0.2757352941176471,
"grad_norm": 11.450309910668999,
"learning_rate": 9.146341463414635e-06,
"loss": 0.53,
"step": 150
},
{
"epoch": 0.29411764705882354,
"grad_norm": 21.677329690439954,
"learning_rate": 9.756097560975611e-06,
"loss": 0.5538,
"step": 160
},
{
"epoch": 0.3125,
"grad_norm": 13.221017712308962,
"learning_rate": 9.999587822598768e-06,
"loss": 0.4616,
"step": 170
},
{
"epoch": 0.33088235294117646,
"grad_norm": 17.10580370679984,
"learning_rate": 9.997069206794246e-06,
"loss": 0.5633,
"step": 180
},
{
"epoch": 0.3492647058823529,
"grad_norm": 12.4225156252439,
"learning_rate": 9.992262114666653e-06,
"loss": 0.6035,
"step": 190
},
{
"epoch": 0.36764705882352944,
"grad_norm": 9.012606555821149,
"learning_rate": 9.985168747689706e-06,
"loss": 0.5822,
"step": 200
},
{
"epoch": 0.3860294117647059,
"grad_norm": 7.876601539287651,
"learning_rate": 9.975792354368019e-06,
"loss": 0.3755,
"step": 210
},
{
"epoch": 0.40441176470588236,
"grad_norm": 6.10882571982772,
"learning_rate": 9.964137228749409e-06,
"loss": 0.4914,
"step": 220
},
{
"epoch": 0.4227941176470588,
"grad_norm": 6.537974128609375,
"learning_rate": 9.950208708458371e-06,
"loss": 0.4604,
"step": 230
},
{
"epoch": 0.4411764705882353,
"grad_norm": 7.108775542718406,
"learning_rate": 9.934013172251654e-06,
"loss": 0.5029,
"step": 240
},
{
"epoch": 0.45955882352941174,
"grad_norm": 13.652621643454376,
"learning_rate": 9.915558037097003e-06,
"loss": 0.5404,
"step": 250
},
{
"epoch": 0.47794117647058826,
"grad_norm": 9.685271320615122,
"learning_rate": 9.894851754776473e-06,
"loss": 0.3555,
"step": 260
},
{
"epoch": 0.4963235294117647,
"grad_norm": 10.504810722270348,
"learning_rate": 9.871903808015812e-06,
"loss": 0.4092,
"step": 270
},
{
"epoch": 0.5147058823529411,
"grad_norm": 12.557835687927717,
"learning_rate": 9.846724706141718e-06,
"loss": 0.4489,
"step": 280
},
{
"epoch": 0.5330882352941176,
"grad_norm": 19.32420837447236,
"learning_rate": 9.819325980268946e-06,
"loss": 0.5297,
"step": 290
},
{
"epoch": 0.5514705882352942,
"grad_norm": 12.56688753149935,
"learning_rate": 9.789720178019483e-06,
"loss": 0.4301,
"step": 300
},
{
"epoch": 0.5698529411764706,
"grad_norm": 8.769304725602016,
"learning_rate": 9.757920857776188e-06,
"loss": 0.5499,
"step": 310
},
{
"epoch": 0.5882352941176471,
"grad_norm": 6.742725850222398,
"learning_rate": 9.723942582473545e-06,
"loss": 0.3723,
"step": 320
},
{
"epoch": 0.6066176470588235,
"grad_norm": 32.81061857477797,
"learning_rate": 9.687800912928362e-06,
"loss": 0.6103,
"step": 330
},
{
"epoch": 0.625,
"grad_norm": 7.250155807815409,
"learning_rate": 9.649512400713497e-06,
"loss": 0.5561,
"step": 340
},
{
"epoch": 0.6433823529411765,
"grad_norm": 8.449295772111478,
"learning_rate": 9.609094580577825e-06,
"loss": 0.4384,
"step": 350
},
{
"epoch": 0.6617647058823529,
"grad_norm": 8.965901929446623,
"learning_rate": 9.566565962415958e-06,
"loss": 0.492,
"step": 360
},
{
"epoch": 0.6801470588235294,
"grad_norm": 10.783592668291883,
"learning_rate": 9.521946022791402e-06,
"loss": 0.5596,
"step": 370
},
{
"epoch": 0.6985294117647058,
"grad_norm": 6.524203722153494,
"learning_rate": 9.475255196016972e-06,
"loss": 0.4657,
"step": 380
},
{
"epoch": 0.7169117647058824,
"grad_norm": 6.844495651307719,
"learning_rate": 9.426514864796648e-06,
"loss": 0.547,
"step": 390
},
{
"epoch": 0.7352941176470589,
"grad_norm": 7.093995062440304,
"learning_rate": 9.375747350433044e-06,
"loss": 0.4821,
"step": 400
},
{
"epoch": 0.7536764705882353,
"grad_norm": 13.008619062718342,
"learning_rate": 9.322975902605082e-06,
"loss": 0.3627,
"step": 410
},
{
"epoch": 0.7720588235294118,
"grad_norm": 7.658314017103231,
"learning_rate": 9.268224688720475e-06,
"loss": 0.2779,
"step": 420
},
{
"epoch": 0.7904411764705882,
"grad_norm": 9.961772688539515,
"learning_rate": 9.211518782847932e-06,
"loss": 0.684,
"step": 430
},
{
"epoch": 0.8088235294117647,
"grad_norm": 5.184448250813704,
"learning_rate": 9.152884154234147e-06,
"loss": 0.3829,
"step": 440
},
{
"epoch": 0.8272058823529411,
"grad_norm": 16.899974205034713,
"learning_rate": 9.092347655410818e-06,
"loss": 0.409,
"step": 450
},
{
"epoch": 0.8455882352941176,
"grad_norm": 9.123647569900998,
"learning_rate": 9.029937009897176e-06,
"loss": 0.4189,
"step": 460
},
{
"epoch": 0.8639705882352942,
"grad_norm": 14.941661911314725,
"learning_rate": 8.96568079950361e-06,
"loss": 0.4846,
"step": 470
},
{
"epoch": 0.8823529411764706,
"grad_norm": 6.576553080941865,
"learning_rate": 8.899608451242233e-06,
"loss": 0.4869,
"step": 480
},
{
"epoch": 0.9007352941176471,
"grad_norm": 9.989273508498824,
"learning_rate": 8.83175022385039e-06,
"loss": 0.368,
"step": 490
},
{
"epoch": 0.9191176470588235,
"grad_norm": 5.168977011450082,
"learning_rate": 8.762137193933241e-06,
"loss": 0.3489,
"step": 500
}
],
"logging_steps": 10,
"max_steps": 1632,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"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": 75890884608.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}