| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.8, |
| "eval_steps": 1, |
| "global_step": 4, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.8, |
| "grad_norm": 15.75, |
| "learning_rate": 5.000000000000001e-07, |
| "loss": 0.9711, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_loss": 1.0638165473937988, |
| "eval_runtime": 30.4688, |
| "eval_samples_per_second": 3.282, |
| "eval_steps_per_second": 0.427, |
| "step": 1 |
| }, |
| { |
| "epoch": 1.4, |
| "grad_norm": 15.8125, |
| "learning_rate": 1.0000000000000002e-06, |
| "loss": 0.9758, |
| "step": 2 |
| }, |
| { |
| "epoch": 1.4, |
| "eval_loss": 1.034848690032959, |
| "eval_runtime": 30.4612, |
| "eval_samples_per_second": 3.283, |
| "eval_steps_per_second": 0.427, |
| "step": 2 |
| }, |
| { |
| "epoch": 2.2, |
| "grad_norm": 13.3125, |
| "learning_rate": 1.5e-06, |
| "loss": 0.9554, |
| "step": 3 |
| }, |
| { |
| "epoch": 2.2, |
| "eval_loss": 0.9725089073181152, |
| "eval_runtime": 30.4647, |
| "eval_samples_per_second": 3.282, |
| "eval_steps_per_second": 0.427, |
| "step": 3 |
| }, |
| { |
| "epoch": 2.8, |
| "grad_norm": 9.75, |
| "learning_rate": 2.0000000000000003e-06, |
| "loss": 0.9379, |
| "step": 4 |
| }, |
| { |
| "epoch": 2.8, |
| "eval_loss": 0.9125163555145264, |
| "eval_runtime": 30.462, |
| "eval_samples_per_second": 3.283, |
| "eval_steps_per_second": 0.427, |
| "step": 4 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 4, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "save_steps": 1, |
| "total_flos": 4.473640512493978e+16, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
