| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.028228652081863093, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005645730416372619, | |
| "eval_loss": 2.1830296516418457, | |
| "eval_runtime": 63.2333, | |
| "eval_samples_per_second": 11.798, | |
| "eval_steps_per_second": 5.899, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0028228652081863093, | |
| "grad_norm": 4.909060001373291, | |
| "learning_rate": 5e-05, | |
| "loss": 8.6593, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0056457304163726185, | |
| "grad_norm": 4.655195713043213, | |
| "learning_rate": 0.0001, | |
| "loss": 9.0873, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0056457304163726185, | |
| "eval_loss": 2.043684720993042, | |
| "eval_runtime": 45.3827, | |
| "eval_samples_per_second": 16.438, | |
| "eval_steps_per_second": 8.219, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008468595624558928, | |
| "grad_norm": 4.215127468109131, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 8.1578, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.011291460832745237, | |
| "grad_norm": 4.801355361938477, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 7.622, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.011291460832745237, | |
| "eval_loss": 1.9020413160324097, | |
| "eval_runtime": 46.7016, | |
| "eval_samples_per_second": 15.974, | |
| "eval_steps_per_second": 7.987, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.014114326040931546, | |
| "grad_norm": 4.917707920074463, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 7.4351, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.016937191249117856, | |
| "grad_norm": 3.883378267288208, | |
| "learning_rate": 5e-05, | |
| "loss": 7.6614, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.016937191249117856, | |
| "eval_loss": 1.8544365167617798, | |
| "eval_runtime": 44.5521, | |
| "eval_samples_per_second": 16.744, | |
| "eval_steps_per_second": 8.372, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.019760056457304165, | |
| "grad_norm": 4.95913028717041, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 7.8261, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.022582921665490474, | |
| "grad_norm": 3.3246023654937744, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 8.0502, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.022582921665490474, | |
| "eval_loss": 1.846505880355835, | |
| "eval_runtime": 45.256, | |
| "eval_samples_per_second": 16.484, | |
| "eval_steps_per_second": 8.242, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.025405786873676783, | |
| "grad_norm": 5.303725719451904, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 8.4275, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.028228652081863093, | |
| "grad_norm": 4.683985233306885, | |
| "learning_rate": 0.0, | |
| "loss": 7.3733, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.028228652081863093, | |
| "eval_loss": 1.842321753501892, | |
| "eval_runtime": 39.8642, | |
| "eval_samples_per_second": 18.714, | |
| "eval_steps_per_second": 9.357, | |
| "step": 50 | |
| } | |
| ], | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 431650504704000.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |