FT_3 / checkpoint-400 /trainer_state.json
sheepy928's picture
Training in progress, step 400, checkpoint
857249c
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.127659574468085,
"eval_steps": 20,
"global_step": 400,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.05,
"learning_rate": 0.0004995563442768412,
"loss": 1.3139,
"step": 10
},
{
"epoch": 0.11,
"learning_rate": 0.0004986690328305235,
"loss": 1.2812,
"step": 20
},
{
"epoch": 0.11,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.8499402403831482,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 125.9807,
"eval_samples_per_second": 11.907,
"eval_steps_per_second": 0.373,
"step": 20
},
{
"epoch": 0.16,
"learning_rate": 0.0004977817213842058,
"loss": 0.832,
"step": 30
},
{
"epoch": 0.21,
"learning_rate": 0.0004968944099378882,
"loss": 0.8156,
"step": 40
},
{
"epoch": 0.21,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.8848057985305786,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 46.3365,
"eval_samples_per_second": 32.372,
"eval_steps_per_second": 1.014,
"step": 40
},
{
"epoch": 0.27,
"learning_rate": 0.0004960070984915705,
"loss": 0.7475,
"step": 50
},
{
"epoch": 0.32,
"learning_rate": 0.0004951197870452529,
"loss": 0.7478,
"step": 60
},
{
"epoch": 0.32,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7691774964332581,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.0551,
"eval_samples_per_second": 247.726,
"eval_steps_per_second": 7.762,
"step": 60
},
{
"epoch": 0.37,
"learning_rate": 0.0004942324755989353,
"loss": 0.7663,
"step": 70
},
{
"epoch": 0.43,
"learning_rate": 0.0004933451641526176,
"loss": 0.6034,
"step": 80
},
{
"epoch": 0.43,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7553095817565918,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.025,
"eval_samples_per_second": 248.962,
"eval_steps_per_second": 7.801,
"step": 80
},
{
"epoch": 0.48,
"learning_rate": 0.0004924578527063,
"loss": 0.9511,
"step": 90
},
{
"epoch": 0.53,
"learning_rate": 0.0004915705412599822,
"loss": 0.7548,
"step": 100
},
{
"epoch": 0.53,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7768574357032776,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 5.9796,
"eval_samples_per_second": 250.851,
"eval_steps_per_second": 7.86,
"step": 100
},
{
"epoch": 0.59,
"learning_rate": 0.0004906832298136646,
"loss": 0.858,
"step": 110
},
{
"epoch": 0.64,
"learning_rate": 0.0004897959183673469,
"loss": 0.8452,
"step": 120
},
{
"epoch": 0.64,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.745326817035675,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.1179,
"eval_samples_per_second": 245.183,
"eval_steps_per_second": 7.682,
"step": 120
},
{
"epoch": 0.69,
"learning_rate": 0.0004889086069210293,
"loss": 0.8152,
"step": 130
},
{
"epoch": 0.74,
"learning_rate": 0.00048802129547471164,
"loss": 0.8972,
"step": 140
},
{
"epoch": 0.74,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7610827088356018,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.0456,
"eval_samples_per_second": 248.113,
"eval_steps_per_second": 7.774,
"step": 140
},
{
"epoch": 0.8,
"learning_rate": 0.000487133984028394,
"loss": 0.7226,
"step": 150
},
{
"epoch": 0.85,
"learning_rate": 0.0004862466725820763,
"loss": 0.9377,
"step": 160
},
{
"epoch": 0.85,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7714303135871887,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 5.9995,
"eval_samples_per_second": 250.022,
"eval_steps_per_second": 7.834,
"step": 160
},
{
"epoch": 0.9,
"learning_rate": 0.0004853593611357587,
"loss": 0.594,
"step": 170
},
{
"epoch": 0.96,
"learning_rate": 0.00048447204968944104,
"loss": 1.0968,
"step": 180
},
{
"epoch": 0.96,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.9954016804695129,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 5.9871,
"eval_samples_per_second": 250.537,
"eval_steps_per_second": 7.85,
"step": 180
},
{
"epoch": 1.01,
"learning_rate": 0.00048358473824312333,
"loss": 0.9644,
"step": 190
},
{
"epoch": 1.06,
"learning_rate": 0.00048269742679680566,
"loss": 0.7502,
"step": 200
},
{
"epoch": 1.06,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7544593214988708,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.0077,
"eval_samples_per_second": 249.679,
"eval_steps_per_second": 7.823,
"step": 200
},
{
"epoch": 1.12,
"learning_rate": 0.000481810115350488,
"loss": 0.7231,
"step": 210
},
{
"epoch": 1.17,
"learning_rate": 0.0004809228039041704,
"loss": 0.7902,
"step": 220
},
{
"epoch": 1.17,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7758567929267883,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.0531,
"eval_samples_per_second": 247.806,
"eval_steps_per_second": 7.765,
"step": 220
},
{
"epoch": 1.22,
"learning_rate": 0.00048003549245785273,
"loss": 0.7591,
"step": 230
},
{
"epoch": 1.28,
"learning_rate": 0.00047914818101153507,
"loss": 0.817,
"step": 240
},
{
"epoch": 1.28,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.767899751663208,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.1711,
"eval_samples_per_second": 243.067,
"eval_steps_per_second": 7.616,
"step": 240
},
{
"epoch": 1.33,
"learning_rate": 0.0004782608695652174,
"loss": 0.6681,
"step": 250
},
{
"epoch": 1.38,
"learning_rate": 0.00047737355811889974,
"loss": 0.9199,
"step": 260
},
{
"epoch": 1.38,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7462519407272339,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.0331,
"eval_samples_per_second": 248.63,
"eval_steps_per_second": 7.79,
"step": 260
},
{
"epoch": 1.44,
"learning_rate": 0.00047648624667258213,
"loss": 0.7409,
"step": 270
},
{
"epoch": 1.49,
"learning_rate": 0.0004755989352262644,
"loss": 0.7246,
"step": 280
},
{
"epoch": 1.49,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7581822872161865,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.1558,
"eval_samples_per_second": 243.673,
"eval_steps_per_second": 7.635,
"step": 280
},
{
"epoch": 1.54,
"learning_rate": 0.00047471162377994675,
"loss": 0.6876,
"step": 290
},
{
"epoch": 1.6,
"learning_rate": 0.0004738243123336291,
"loss": 0.7508,
"step": 300
},
{
"epoch": 1.6,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7464602589607239,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.1548,
"eval_samples_per_second": 243.711,
"eval_steps_per_second": 7.636,
"step": 300
},
{
"epoch": 1.65,
"learning_rate": 0.0004729370008873114,
"loss": 0.8539,
"step": 310
},
{
"epoch": 1.7,
"learning_rate": 0.0004720496894409938,
"loss": 0.7011,
"step": 320
},
{
"epoch": 1.7,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.788334310054779,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.1655,
"eval_samples_per_second": 243.289,
"eval_steps_per_second": 7.623,
"step": 320
},
{
"epoch": 1.76,
"learning_rate": 0.00047116237799467615,
"loss": 0.9111,
"step": 330
},
{
"epoch": 1.81,
"learning_rate": 0.0004702750665483585,
"loss": 0.8442,
"step": 340
},
{
"epoch": 1.81,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7656621336936951,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.2177,
"eval_samples_per_second": 241.249,
"eval_steps_per_second": 7.559,
"step": 340
},
{
"epoch": 1.86,
"learning_rate": 0.00046938775510204083,
"loss": 0.7179,
"step": 350
},
{
"epoch": 1.91,
"learning_rate": 0.00046850044365572317,
"loss": 1.0021,
"step": 360
},
{
"epoch": 1.91,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7755117416381836,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.0871,
"eval_samples_per_second": 246.422,
"eval_steps_per_second": 7.721,
"step": 360
},
{
"epoch": 1.97,
"learning_rate": 0.0004676131322094055,
"loss": 0.8331,
"step": 370
},
{
"epoch": 2.02,
"learning_rate": 0.00046672582076308784,
"loss": 0.7982,
"step": 380
},
{
"epoch": 2.02,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.8436231017112732,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.0719,
"eval_samples_per_second": 247.04,
"eval_steps_per_second": 7.741,
"step": 380
},
{
"epoch": 2.07,
"learning_rate": 0.0004658385093167702,
"loss": 0.8408,
"step": 390
},
{
"epoch": 2.13,
"learning_rate": 0.0004649511978704525,
"loss": 0.8299,
"step": 400
},
{
"epoch": 2.13,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7582988142967224,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.2274,
"eval_samples_per_second": 240.873,
"eval_steps_per_second": 7.547,
"step": 400
}
],
"logging_steps": 10,
"max_steps": 5640,
"num_train_epochs": 30,
"save_steps": 100,
"total_flos": 839857999970304.0,
"trial_name": null,
"trial_params": null
}