| { | |
| "best_metric": 2.4199516773223877, | |
| "best_model_checkpoint": "./outputs/checkpoint-700", | |
| "epoch": 0.5100182149362478, | |
| "eval_steps": 100, | |
| "global_step": 700, | |
| "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": 2.7205, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.6170873641967773, | |
| "eval_runtime": 228.505, | |
| "eval_samples_per_second": 27.457, | |
| "eval_steps_per_second": 3.435, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5807, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.5663836002349854, | |
| "eval_runtime": 210.5353, | |
| "eval_samples_per_second": 29.8, | |
| "eval_steps_per_second": 3.729, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5398, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.531459331512451, | |
| "eval_runtime": 245.5646, | |
| "eval_samples_per_second": 25.549, | |
| "eval_steps_per_second": 3.197, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5133, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 2.497565746307373, | |
| "eval_runtime": 244.6469, | |
| "eval_samples_per_second": 25.645, | |
| "eval_steps_per_second": 3.209, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4697, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 2.4726672172546387, | |
| "eval_runtime": 244.9375, | |
| "eval_samples_per_second": 25.615, | |
| "eval_steps_per_second": 3.205, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4435, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 2.4451191425323486, | |
| "eval_runtime": 216.439, | |
| "eval_samples_per_second": 28.987, | |
| "eval_steps_per_second": 3.627, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 2.426, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 2.4199516773223877, | |
| "eval_runtime": 206.0859, | |
| "eval_samples_per_second": 30.444, | |
| "eval_steps_per_second": 3.809, | |
| "step": 700 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 2.055883980423168e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |