| { | |
| "best_metric": 1.8804633617401123, | |
| "best_model_checkpoint": "./outputs/checkpoint-600", | |
| "epoch": 0.4371584699453552, | |
| "eval_steps": 100, | |
| "global_step": 600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3152, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.160670518875122, | |
| "eval_runtime": 143.4488, | |
| "eval_samples_per_second": 43.737, | |
| "eval_steps_per_second": 5.472, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1094, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.0639851093292236, | |
| "eval_runtime": 143.5664, | |
| "eval_samples_per_second": 43.701, | |
| "eval_steps_per_second": 5.468, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0347, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.0028491020202637, | |
| "eval_runtime": 143.5409, | |
| "eval_samples_per_second": 43.709, | |
| "eval_steps_per_second": 5.469, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9808, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.9539737701416016, | |
| "eval_runtime": 143.5859, | |
| "eval_samples_per_second": 43.695, | |
| "eval_steps_per_second": 5.467, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9262, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.911293625831604, | |
| "eval_runtime": 143.5587, | |
| "eval_samples_per_second": 43.703, | |
| "eval_steps_per_second": 5.468, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8974, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.8804633617401123, | |
| "eval_runtime": 143.5407, | |
| "eval_samples_per_second": 43.709, | |
| "eval_steps_per_second": 5.469, | |
| "step": 600 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 3.576193971702989e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |