Output_llama2_70-15-15 / trainer_state.json
Ahatsham's picture
Model save
333f7cc verified
{
"best_metric": 0.6282051282051282,
"best_model_checkpoint": "Output_llama2_70-15-15/checkpoint-874",
"epoch": 20.0,
"eval_steps": 500,
"global_step": 920,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_accuracy": 0.5641025641025641,
"eval_balanced_accuracy": 0.5764176417641764,
"eval_loss": 0.7111307978630066,
"eval_runtime": 111.2436,
"eval_samples_per_second": 1.402,
"eval_steps_per_second": 0.09,
"step": 46
},
{
"epoch": 2.0,
"eval_accuracy": 0.5576923076923077,
"eval_balanced_accuracy": 0.565625,
"eval_loss": 0.7042574882507324,
"eval_runtime": 111.3987,
"eval_samples_per_second": 1.4,
"eval_steps_per_second": 0.09,
"step": 92
},
{
"epoch": 3.0,
"eval_accuracy": 0.5192307692307693,
"eval_balanced_accuracy": 0.5141509433962264,
"eval_loss": 0.6619334816932678,
"eval_runtime": 111.2028,
"eval_samples_per_second": 1.403,
"eval_steps_per_second": 0.09,
"step": 138
},
{
"epoch": 4.0,
"eval_accuracy": 0.5512820512820513,
"eval_balanced_accuracy": 0.5594967674296698,
"eval_loss": 0.7012639045715332,
"eval_runtime": 111.1032,
"eval_samples_per_second": 1.404,
"eval_steps_per_second": 0.09,
"step": 184
},
{
"epoch": 5.0,
"eval_accuracy": 0.5576923076923077,
"eval_balanced_accuracy": 0.5620300751879699,
"eval_loss": 0.6493321657180786,
"eval_runtime": 111.1592,
"eval_samples_per_second": 1.403,
"eval_steps_per_second": 0.09,
"step": 230
},
{
"epoch": 6.0,
"eval_accuracy": 0.5641025641025641,
"eval_balanced_accuracy": 0.5670505563146594,
"eval_loss": 0.6495689749717712,
"eval_runtime": 111.1491,
"eval_samples_per_second": 1.404,
"eval_steps_per_second": 0.09,
"step": 276
},
{
"epoch": 7.0,
"eval_accuracy": 0.5769230769230769,
"eval_balanced_accuracy": 0.5798319327731092,
"eval_loss": 0.6466049551963806,
"eval_runtime": 111.1355,
"eval_samples_per_second": 1.404,
"eval_steps_per_second": 0.09,
"step": 322
},
{
"epoch": 8.0,
"eval_accuracy": 0.5512820512820513,
"eval_balanced_accuracy": 0.5527314738405678,
"eval_loss": 0.6747537851333618,
"eval_runtime": 111.3484,
"eval_samples_per_second": 1.401,
"eval_steps_per_second": 0.09,
"step": 368
},
{
"epoch": 9.0,
"eval_accuracy": 0.5705128205128205,
"eval_balanced_accuracy": 0.5691844919786097,
"eval_loss": 0.6550834774971008,
"eval_runtime": 111.4062,
"eval_samples_per_second": 1.4,
"eval_steps_per_second": 0.09,
"step": 414
},
{
"epoch": 10.0,
"eval_accuracy": 0.5833333333333334,
"eval_balanced_accuracy": 0.6063162970106075,
"eval_loss": 0.6205254793167114,
"eval_runtime": 111.3582,
"eval_samples_per_second": 1.401,
"eval_steps_per_second": 0.09,
"step": 460
},
{
"epoch": 10.869565217391305,
"grad_norm": 3.162184715270996,
"learning_rate": 4.565217391304348e-06,
"loss": 0.6541,
"step": 500
},
{
"epoch": 11.0,
"eval_accuracy": 0.6025641025641025,
"eval_balanced_accuracy": 0.6019753086419752,
"eval_loss": 0.6536551117897034,
"eval_runtime": 111.1486,
"eval_samples_per_second": 1.404,
"eval_steps_per_second": 0.09,
"step": 506
},
{
"epoch": 12.0,
"eval_accuracy": 0.6153846153846154,
"eval_balanced_accuracy": 0.6167385677308024,
"eval_loss": 0.6378962397575378,
"eval_runtime": 111.1235,
"eval_samples_per_second": 1.404,
"eval_steps_per_second": 0.09,
"step": 552
},
{
"epoch": 13.0,
"eval_accuracy": 0.6025641025641025,
"eval_balanced_accuracy": 0.6107190539767309,
"eval_loss": 0.6242974400520325,
"eval_runtime": 111.2048,
"eval_samples_per_second": 1.403,
"eval_steps_per_second": 0.09,
"step": 598
},
{
"epoch": 14.0,
"eval_accuracy": 0.6025641025641025,
"eval_balanced_accuracy": 0.60743725957135,
"eval_loss": 0.6248365640640259,
"eval_runtime": 111.2066,
"eval_samples_per_second": 1.403,
"eval_steps_per_second": 0.09,
"step": 644
},
{
"epoch": 15.0,
"eval_accuracy": 0.6217948717948718,
"eval_balanced_accuracy": 0.6369565217391304,
"eval_loss": 0.6171802878379822,
"eval_runtime": 111.2635,
"eval_samples_per_second": 1.402,
"eval_steps_per_second": 0.09,
"step": 690
},
{
"epoch": 16.0,
"eval_accuracy": 0.6153846153846154,
"eval_balanced_accuracy": 0.6201620162016201,
"eval_loss": 0.6236566305160522,
"eval_runtime": 111.2833,
"eval_samples_per_second": 1.402,
"eval_steps_per_second": 0.09,
"step": 736
},
{
"epoch": 17.0,
"eval_accuracy": 0.6217948717948718,
"eval_balanced_accuracy": 0.6230267673301304,
"eval_loss": 0.630794107913971,
"eval_runtime": 111.1945,
"eval_samples_per_second": 1.403,
"eval_steps_per_second": 0.09,
"step": 782
},
{
"epoch": 18.0,
"eval_accuracy": 0.6217948717948718,
"eval_balanced_accuracy": 0.6318867924528302,
"eval_loss": 0.6179038882255554,
"eval_runtime": 111.4097,
"eval_samples_per_second": 1.4,
"eval_steps_per_second": 0.09,
"step": 828
},
{
"epoch": 19.0,
"eval_accuracy": 0.6282051282051282,
"eval_balanced_accuracy": 0.63264221158958,
"eval_loss": 0.6252104640007019,
"eval_runtime": 111.3148,
"eval_samples_per_second": 1.401,
"eval_steps_per_second": 0.09,
"step": 874
},
{
"epoch": 20.0,
"eval_accuracy": 0.6282051282051282,
"eval_balanced_accuracy": 0.63264221158958,
"eval_loss": 0.6249805688858032,
"eval_runtime": 111.384,
"eval_samples_per_second": 1.401,
"eval_steps_per_second": 0.09,
"step": 920
},
{
"epoch": 20.0,
"step": 920,
"total_flos": 4.970048339828736e+17,
"train_loss": 0.6203255363132643,
"train_runtime": 33373.6734,
"train_samples_per_second": 0.437,
"train_steps_per_second": 0.028
}
],
"logging_steps": 500,
"max_steps": 920,
"num_input_tokens_seen": 0,
"num_train_epochs": 20,
"save_steps": 500,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 19,
"early_stopping_threshold": 0.0
},
"attributes": {
"early_stopping_patience_counter": 0
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 4.970048339828736e+17,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}