| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0012827048627670247, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.288986827607755e-05, | |
| "eval_loss": 2.7174880504608154, | |
| "eval_runtime": 2021.0944, | |
| "eval_samples_per_second": 6.334, | |
| "eval_steps_per_second": 3.167, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00016444934138038777, | |
| "grad_norm": 2.2541286945343018, | |
| "learning_rate": 5e-05, | |
| "loss": 2.8779, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00032889868276077554, | |
| "grad_norm": 3.0431084632873535, | |
| "learning_rate": 0.0001, | |
| "loss": 2.5379, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00032889868276077554, | |
| "eval_loss": 2.3707923889160156, | |
| "eval_runtime": 2026.1498, | |
| "eval_samples_per_second": 6.318, | |
| "eval_steps_per_second": 3.159, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0004933480241411634, | |
| "grad_norm": 3.701636791229248, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.1496, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0006577973655215511, | |
| "grad_norm": 3.3653860092163086, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.8656, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0006577973655215511, | |
| "eval_loss": 1.9287056922912598, | |
| "eval_runtime": 2025.7457, | |
| "eval_samples_per_second": 6.32, | |
| "eval_steps_per_second": 3.16, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0008222467069019389, | |
| "grad_norm": 3.4819860458374023, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.8805, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0009866960482823267, | |
| "grad_norm": 3.7152276039123535, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9245, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0009866960482823267, | |
| "eval_loss": 1.8017830848693848, | |
| "eval_runtime": 2023.9174, | |
| "eval_samples_per_second": 6.325, | |
| "eval_steps_per_second": 3.163, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0011511453896627143, | |
| "grad_norm": 3.436457633972168, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.626, | |
| "step": 35 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.490651304689664e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |