| { | |
| "best_metric": 1.9188886880874634, | |
| "best_model_checkpoint": "gen_test/checkpoint-3126", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 3126, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 167490.5625, | |
| "learning_rate": 5.520153550863724e-05, | |
| "loss": 2.0195, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 149414.109375, | |
| "learning_rate": 5.040307101727447e-05, | |
| "loss": 1.8734, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 117651.6953125, | |
| "learning_rate": 4.560460652591171e-05, | |
| "loss": 1.8297, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.9658416509628296, | |
| "eval_runtime": 56.3883, | |
| "eval_samples_per_second": 177.342, | |
| "eval_steps_per_second": 2.784, | |
| "step": 1563 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 112915.5, | |
| "learning_rate": 4.0806142034548946e-05, | |
| "loss": 1.7959, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 105164.921875, | |
| "learning_rate": 3.6007677543186184e-05, | |
| "loss": 1.7747, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 94653.9765625, | |
| "learning_rate": 3.1209213051823415e-05, | |
| "loss": 1.7648, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.9188886880874634, | |
| "eval_runtime": 56.4098, | |
| "eval_samples_per_second": 177.274, | |
| "eval_steps_per_second": 2.783, | |
| "step": 3126 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6252, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "total_flos": 1.045168128e+17, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |