| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.571428571428571, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 1.5, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 2.755, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "eval_loss": 3.1059346199035645, | |
| "eval_runtime": 0.5642, | |
| "eval_samples_per_second": 17.725, | |
| "eval_steps_per_second": 1.772, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 1.34375, | |
| "learning_rate": 0.00019730448705798239, | |
| "loss": 2.6986, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.8355600833892822, | |
| "eval_runtime": 0.5011, | |
| "eval_samples_per_second": 19.954, | |
| "eval_steps_per_second": 1.995, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 2.375, | |
| "learning_rate": 0.0001686241637868734, | |
| "loss": 2.2478, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_loss": 2.730991840362549, | |
| "eval_runtime": 0.6312, | |
| "eval_samples_per_second": 15.842, | |
| "eval_steps_per_second": 1.584, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.6772868633270264, | |
| "eval_runtime": 0.5219, | |
| "eval_samples_per_second": 19.162, | |
| "eval_steps_per_second": 1.916, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 1.9892, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "eval_loss": 2.648698091506958, | |
| "eval_runtime": 0.6365, | |
| "eval_samples_per_second": 15.712, | |
| "eval_steps_per_second": 1.571, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.6015625, | |
| "learning_rate": 6.039202339608432e-05, | |
| "loss": 1.8401, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.618410348892212, | |
| "eval_runtime": 0.5361, | |
| "eval_samples_per_second": 18.654, | |
| "eval_steps_per_second": 1.865, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_loss": 2.6133430004119873, | |
| "eval_runtime": 0.6782, | |
| "eval_samples_per_second": 14.744, | |
| "eval_steps_per_second": 1.474, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 7.142857142857143, | |
| "grad_norm": 0.859375, | |
| "learning_rate": 1.6451218858706374e-05, | |
| "loss": 1.7508, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.6098170280456543, | |
| "eval_runtime": 0.5541, | |
| "eval_samples_per_second": 18.047, | |
| "eval_steps_per_second": 1.805, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 0.52734375, | |
| "learning_rate": 0.0, | |
| "loss": 1.7163, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_loss": 2.6123337745666504, | |
| "eval_runtime": 0.5428, | |
| "eval_samples_per_second": 18.424, | |
| "eval_steps_per_second": 1.842, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "step": 30, | |
| "total_flos": 1.762120022622208e+16, | |
| "train_loss": 2.042347224553426, | |
| "train_runtime": 106.802, | |
| "train_samples_per_second": 14.232, | |
| "train_steps_per_second": 0.281 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "total_flos": 1.762120022622208e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |