| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9931662870159453, | |
| "eval_steps": 500, | |
| "global_step": 109, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.009111617312072893, | |
| "grad_norm": 2847.998104709306, | |
| "learning_rate": 1.8181818181818183e-06, | |
| "loss": 42.2184, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04555808656036447, | |
| "grad_norm": 562.0305390689821, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 29.7876, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.09111617312072894, | |
| "grad_norm": 94.49823214647388, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 18.8302, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1366742596810934, | |
| "grad_norm": 54.33888379443643, | |
| "learning_rate": 1.991790013823246e-05, | |
| "loss": 7.0598, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.18223234624145787, | |
| "grad_norm": 24.89922969697027, | |
| "learning_rate": 1.9586678530366607e-05, | |
| "loss": 2.67, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.22779043280182232, | |
| "grad_norm": 95.75519440146691, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 2.059, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2733485193621868, | |
| "grad_norm": 37.044831894790526, | |
| "learning_rate": 1.820172254596956e-05, | |
| "loss": 1.7679, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.31890660592255127, | |
| "grad_norm": 36.79419087973788, | |
| "learning_rate": 1.7183493500977277e-05, | |
| "loss": 1.5439, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.36446469248291574, | |
| "grad_norm": 29.940404675815028, | |
| "learning_rate": 1.598110530491216e-05, | |
| "loss": 1.4442, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.41002277904328016, | |
| "grad_norm": 4.190592272065602, | |
| "learning_rate": 1.4625382902408356e-05, | |
| "loss": 1.2511, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.45558086560364464, | |
| "grad_norm": 9.541987280500265, | |
| "learning_rate": 1.315108218023621e-05, | |
| "loss": 1.2099, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5011389521640092, | |
| "grad_norm": 6.4393691976336305, | |
| "learning_rate": 1.1595998950333794e-05, | |
| "loss": 1.0664, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.5466970387243736, | |
| "grad_norm": 3.8451292577941585, | |
| "learning_rate": 1e-05, | |
| "loss": 0.968, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.592255125284738, | |
| "grad_norm": 2.4425581326450616, | |
| "learning_rate": 8.404001049666211e-06, | |
| "loss": 0.9188, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.6378132118451025, | |
| "grad_norm": 2.7821675174035865, | |
| "learning_rate": 6.848917819763794e-06, | |
| "loss": 0.881, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.683371298405467, | |
| "grad_norm": 4.6204696647361665, | |
| "learning_rate": 5.37461709759165e-06, | |
| "loss": 0.8782, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.7289293849658315, | |
| "grad_norm": 2.8829796033312296, | |
| "learning_rate": 4.01889469508784e-06, | |
| "loss": 0.8517, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.7744874715261959, | |
| "grad_norm": 1.6103793214527409, | |
| "learning_rate": 2.8165064990227255e-06, | |
| "loss": 0.8213, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.8200455580865603, | |
| "grad_norm": 1.3054680157700491, | |
| "learning_rate": 1.7982774540304404e-06, | |
| "loss": 0.8088, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8656036446469249, | |
| "grad_norm": 1.0464961249658165, | |
| "learning_rate": 9.903113209758098e-07, | |
| "loss": 0.7988, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.9111617312072893, | |
| "grad_norm": 0.8824867589954551, | |
| "learning_rate": 4.133214696333943e-07, | |
| "loss": 0.7836, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9567198177676538, | |
| "grad_norm": 0.8602109237194131, | |
| "learning_rate": 8.209986176753947e-08, | |
| "loss": 0.7827, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.9931662870159453, | |
| "eval_loss": 2.7200307846069336, | |
| "eval_runtime": 0.4579, | |
| "eval_samples_per_second": 45.861, | |
| "eval_steps_per_second": 2.184, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 0.9931662870159453, | |
| "step": 109, | |
| "total_flos": 14975678545920.0, | |
| "train_loss": 3.682411685996099, | |
| "train_runtime": 758.7359, | |
| "train_samples_per_second": 37.012, | |
| "train_steps_per_second": 0.144 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 109, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 14975678545920.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |