| { | |
| "best_metric": 1.716844081878662, | |
| "best_model_checkpoint": "output/nervy/checkpoint-67", | |
| "epoch": 1.0, | |
| "global_step": 67, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00013532330330528217, | |
| "loss": 2.4761, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00012979589515943672, | |
| "loss": 2.0363, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0001209202032183722, | |
| "loss": 2.0351, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00010918185377156764, | |
| "loss": 1.8591, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 9.522310109331633e-05, | |
| "loss": 1.6691, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 7.980768702075116e-05, | |
| "loss": 1.7511, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 6.377905343835293e-05, | |
| "loss": 1.8029, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.8014194038976637e-05, | |
| "loss": 1.8511, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 3.337567032488183e-05, | |
| "loss": 1.7338, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 2.0664417254371463e-05, | |
| "loss": 1.9426, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.0575920734825916e-05, | |
| "loss": 1.8186, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 3.66216467507622e-06, | |
| "loss": 1.9079, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 3.014296332977907e-07, | |
| "loss": 1.8372, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.716844081878662, | |
| "eval_runtime": 2.1609, | |
| "eval_samples_per_second": 41.187, | |
| "eval_steps_per_second": 5.553, | |
| "step": 67 | |
| } | |
| ], | |
| "max_steps": 67, | |
| "num_train_epochs": 1, | |
| "total_flos": 69764972544000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |