| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 21.60628890991211, | |
| "learning_rate": 2e-05, | |
| "loss": 1.2546, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 21.794620513916016, | |
| "learning_rate": 1.9863613034027224e-05, | |
| "loss": 1.2546, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 13.444245338439941, | |
| "learning_rate": 1.9458172417006347e-05, | |
| "loss": 1.1815, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 9.568578720092773, | |
| "learning_rate": 1.879473751206489e-05, | |
| "loss": 1.0718, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 8.503974914550781, | |
| "learning_rate": 1.789140509396394e-05, | |
| "loss": 0.9886, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9387755102040817, | |
| "eval_accuracy_<|content|>": 0.0, | |
| "eval_accuracy_<|from|>": -1, | |
| "eval_accuracy_<|recipient|>": -1, | |
| "eval_accuracy_<|stop|>": 0.0, | |
| "eval_accuracy_first_token": 1.0, | |
| "eval_accuracy_first_token_get": 1.0, | |
| "eval_accuracy_first_token_get_total": 2, | |
| "eval_accuracy_total_num_<|content|>": 2, | |
| "eval_accuracy_total_num_<|from|>": 0, | |
| "eval_accuracy_total_num_<|recipient|>": 0, | |
| "eval_accuracy_total_num_<|stop|>": 2, | |
| "eval_first_token_param_values": 0.9, | |
| "eval_first_token_param_values_total": 10, | |
| "eval_loss": 0.897951602935791, | |
| "eval_perplexity": 1.0108013242189644, | |
| "eval_runtime": 0.885, | |
| "eval_samples_per_second": 1.13, | |
| "eval_steps_per_second": 1.13, | |
| "eval_total_number_first_token": 2, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 8.138020515441895, | |
| "learning_rate": 1.6772815716257414e-05, | |
| "loss": 0.8967, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 8.324158668518066, | |
| "learning_rate": 1.5469481581224274e-05, | |
| "loss": 0.8374, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 8.041045188903809, | |
| "learning_rate": 1.4016954246529697e-05, | |
| "loss": 0.7753, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 7.898632526397705, | |
| "learning_rate": 1.2454854871407993e-05, | |
| "loss": 0.7203, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 6.87092924118042, | |
| "learning_rate": 1.0825793454723325e-05, | |
| "loss": 0.6677, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9489795918367347, | |
| "eval_accuracy_<|content|>": 0.0, | |
| "eval_accuracy_<|from|>": -1, | |
| "eval_accuracy_<|recipient|>": -1, | |
| "eval_accuracy_<|stop|>": 0.0, | |
| "eval_accuracy_first_token": 1.0, | |
| "eval_accuracy_first_token_get": 1.0, | |
| "eval_accuracy_first_token_get_total": 2, | |
| "eval_accuracy_total_num_<|content|>": 2, | |
| "eval_accuracy_total_num_<|from|>": 0, | |
| "eval_accuracy_total_num_<|recipient|>": 0, | |
| "eval_accuracy_total_num_<|stop|>": 2, | |
| "eval_first_token_param_values": 0.9, | |
| "eval_first_token_param_values_total": 10, | |
| "eval_loss": 0.627693235874176, | |
| "eval_perplexity": 1.0075382125074093, | |
| "eval_runtime": 0.878, | |
| "eval_samples_per_second": 1.139, | |
| "eval_steps_per_second": 1.139, | |
| "eval_total_number_first_token": 2, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 6.593441009521484, | |
| "learning_rate": 9.174206545276678e-06, | |
| "loss": 0.6275, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 6.028723239898682, | |
| "learning_rate": 7.545145128592009e-06, | |
| "loss": 0.5971, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 5.46033239364624, | |
| "learning_rate": 5.983045753470308e-06, | |
| "loss": 0.5745, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 4.930461883544922, | |
| "learning_rate": 4.530518418775734e-06, | |
| "loss": 0.5558, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 4.64890193939209, | |
| "learning_rate": 3.2271842837425917e-06, | |
| "loss": 0.5455, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9489795918367347, | |
| "eval_accuracy_<|content|>": 0.0, | |
| "eval_accuracy_<|from|>": -1, | |
| "eval_accuracy_<|recipient|>": -1, | |
| "eval_accuracy_<|stop|>": 0.0, | |
| "eval_accuracy_first_token": 1.0, | |
| "eval_accuracy_first_token_get": 1.0, | |
| "eval_accuracy_first_token_get_total": 2, | |
| "eval_accuracy_total_num_<|content|>": 2, | |
| "eval_accuracy_total_num_<|from|>": 0, | |
| "eval_accuracy_total_num_<|recipient|>": 0, | |
| "eval_accuracy_total_num_<|stop|>": 2, | |
| "eval_first_token_param_values": 0.9, | |
| "eval_first_token_param_values_total": 10, | |
| "eval_loss": 0.5380075573921204, | |
| "eval_perplexity": 1.0064576742544533, | |
| "eval_runtime": 0.8779, | |
| "eval_samples_per_second": 1.139, | |
| "eval_steps_per_second": 1.139, | |
| "eval_total_number_first_token": 2, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 4.5582594871521, | |
| "learning_rate": 2.1085949060360654e-06, | |
| "loss": 0.5374, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 4.519166946411133, | |
| "learning_rate": 1.2052624879351105e-06, | |
| "loss": 0.5316, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 4.582608699798584, | |
| "learning_rate": 5.418275829936537e-07, | |
| "loss": 0.5291, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.5061774253845215, | |
| "learning_rate": 1.3638696597277678e-07, | |
| "loss": 0.5265, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 4.478773593902588, | |
| "learning_rate": 0.0, | |
| "loss": 0.5249, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9489795918367347, | |
| "eval_accuracy_<|content|>": 0.0, | |
| "eval_accuracy_<|from|>": -1, | |
| "eval_accuracy_<|recipient|>": -1, | |
| "eval_accuracy_<|stop|>": 0.0, | |
| "eval_accuracy_first_token": 1.0, | |
| "eval_accuracy_first_token_get": 1.0, | |
| "eval_accuracy_first_token_get_total": 2, | |
| "eval_accuracy_total_num_<|content|>": 2, | |
| "eval_accuracy_total_num_<|from|>": 0, | |
| "eval_accuracy_total_num_<|recipient|>": 0, | |
| "eval_accuracy_total_num_<|stop|>": 2, | |
| "eval_first_token_param_values": 0.9, | |
| "eval_first_token_param_values_total": 10, | |
| "eval_loss": 0.5260699987411499, | |
| "eval_perplexity": 1.0063139350239472, | |
| "eval_runtime": 0.8774, | |
| "eval_samples_per_second": 1.14, | |
| "eval_steps_per_second": 1.14, | |
| "eval_total_number_first_token": 2, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 20, | |
| "total_flos": 7935331915530240.0, | |
| "train_loss": 0.7599329292774201, | |
| "train_runtime": 382.8027, | |
| "train_samples_per_second": 0.052, | |
| "train_steps_per_second": 0.052 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7935331915530240.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |