| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.3787398317937405, | |
| "eval_steps": 2000, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1.200600300150075e-06, | |
| "loss": 0.8809, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 0.7049718499183655, | |
| "eval_runtime": 2269.7212, | |
| "eval_samples_per_second": 1.37, | |
| "eval_steps_per_second": 0.171, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 9.004502251125563e-07, | |
| "loss": 0.6722, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 0.6513694524765015, | |
| "eval_runtime": 2273.8985, | |
| "eval_samples_per_second": 1.367, | |
| "eval_steps_per_second": 0.171, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 6.003001500750375e-07, | |
| "loss": 0.6416, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "eval_loss": 0.6299079656600952, | |
| "eval_runtime": 2271.2854, | |
| "eval_samples_per_second": 1.369, | |
| "eval_steps_per_second": 0.171, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 3.0015007503751875e-07, | |
| "loss": 0.6223, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 0.620364785194397, | |
| "eval_runtime": 2271.0043, | |
| "eval_samples_per_second": 1.369, | |
| "eval_steps_per_second": 0.171, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0, | |
| "loss": 0.6103, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 0.6169580221176147, | |
| "eval_runtime": 2271.2568, | |
| "eval_samples_per_second": 1.369, | |
| "eval_steps_per_second": 0.171, | |
| "step": 10000 | |
| } | |
| ], | |
| "logging_steps": 2000, | |
| "max_steps": 10000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 2000, | |
| "total_flos": 3.3548553486336e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |