| { |
| "best_metric": 0.30574527382850647, |
| "best_model_checkpoint": "./results/run-0/checkpoint-2000", |
| "epoch": 2.0, |
| "eval_steps": 500, |
| "global_step": 2000, |
| "is_hyper_param_search": true, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.5, |
| "grad_norm": 4.100419521331787, |
| "learning_rate": 3.1570541968021056e-06, |
| "loss": 1.3549, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 11.960673332214355, |
| "learning_rate": 2.8062703971574272e-06, |
| "loss": 0.8349, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7905, |
| "eval_loss": 0.6183840036392212, |
| "eval_runtime": 9.0459, |
| "eval_samples_per_second": 221.096, |
| "eval_steps_per_second": 27.637, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 5.748434543609619, |
| "learning_rate": 2.4554865975127484e-06, |
| "loss": 0.5184, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 22.25712776184082, |
| "learning_rate": 2.10470279786807e-06, |
| "loss": 0.384, |
| "step": 2000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.909, |
| "eval_loss": 0.30574527382850647, |
| "eval_runtime": 9.0588, |
| "eval_samples_per_second": 220.779, |
| "eval_steps_per_second": 27.597, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 5000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "total_flos": 4239259140096000.0, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": { |
| "learning_rate": 3.507837996446784e-06, |
| "num_train_epochs": 5, |
| "per_device_train_batch_size": 16, |
| "seed": 16 |
| } |
| } |
|
|
