|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 4.0, |
|
"eval_steps": 500, |
|
"global_step": 750, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.5333333333333333, |
|
"grad_norm": 0.275871217250824, |
|
"learning_rate": 3e-05, |
|
"loss": 1.8545, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.9973333333333333, |
|
"eval_accuracy": 0.6037399103139014, |
|
"eval_loss": 1.7123007774353027, |
|
"eval_runtime": 6.2976, |
|
"eval_samples_per_second": 79.395, |
|
"eval_steps_per_second": 10.004, |
|
"step": 187 |
|
}, |
|
{ |
|
"epoch": 0.9973333333333333, |
|
"eval_exact_match": 14.4, |
|
"eval_f1": 22.843015873015876, |
|
"eval_qa_bleu": 9.471299126615676, |
|
"eval_qa_exact_match": 0.112, |
|
"eval_recite_bleu": 12.267251092702137, |
|
"eval_recite_exact_match": 0.0, |
|
"step": 187 |
|
}, |
|
{ |
|
"epoch": 1.0666666666666667, |
|
"grad_norm": 0.32765188813209534, |
|
"learning_rate": 3e-05, |
|
"loss": 1.7408, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"grad_norm": 0.3532409369945526, |
|
"learning_rate": 3e-05, |
|
"loss": 1.7063, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.6058565022421525, |
|
"eval_loss": 1.69453763961792, |
|
"eval_runtime": 6.5928, |
|
"eval_samples_per_second": 75.841, |
|
"eval_steps_per_second": 9.556, |
|
"step": 375 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_exact_match": 14.4, |
|
"eval_f1": 23.29731601731603, |
|
"eval_qa_bleu": 9.4408132575913, |
|
"eval_qa_exact_match": 0.11, |
|
"eval_recite_bleu": 13.027794116434837, |
|
"eval_recite_exact_match": 0.0, |
|
"step": 375 |
|
}, |
|
{ |
|
"epoch": 2.1333333333333333, |
|
"grad_norm": 0.3777780532836914, |
|
"learning_rate": 3e-05, |
|
"loss": 1.7094, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 2.6666666666666665, |
|
"grad_norm": 0.3971036374568939, |
|
"learning_rate": 3e-05, |
|
"loss": 1.6702, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.997333333333333, |
|
"eval_accuracy": 0.6065919282511211, |
|
"eval_loss": 1.6851236820220947, |
|
"eval_runtime": 6.3104, |
|
"eval_samples_per_second": 79.234, |
|
"eval_steps_per_second": 9.984, |
|
"step": 562 |
|
}, |
|
{ |
|
"epoch": 2.997333333333333, |
|
"eval_exact_match": 14.0, |
|
"eval_f1": 23.306825396825417, |
|
"eval_qa_bleu": 8.683248465366908, |
|
"eval_qa_exact_match": 0.11, |
|
"eval_recite_bleu": 12.39336052958371, |
|
"eval_recite_exact_match": 0.006, |
|
"step": 562 |
|
}, |
|
{ |
|
"epoch": 3.2, |
|
"grad_norm": 0.4782978594303131, |
|
"learning_rate": 3e-05, |
|
"loss": 1.6574, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 3.7333333333333334, |
|
"grad_norm": 0.52582186460495, |
|
"learning_rate": 3e-05, |
|
"loss": 1.6356, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.6074977578475337, |
|
"eval_loss": 1.682567834854126, |
|
"eval_runtime": 6.65, |
|
"eval_samples_per_second": 75.188, |
|
"eval_steps_per_second": 9.474, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_exact_match": 15.0, |
|
"eval_f1": 23.366825396825416, |
|
"eval_qa_bleu": 9.05994637010322, |
|
"eval_qa_exact_match": 0.114, |
|
"eval_recite_bleu": 13.060415058384661, |
|
"eval_recite_exact_match": 0.006, |
|
"step": 750 |
|
} |
|
], |
|
"logging_steps": 100, |
|
"max_steps": 9350, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 50, |
|
"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": 8.267771833168691e+16, |
|
"train_batch_size": 2, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|