| { | |
| "best_metric": 1.7112771272659302, | |
| "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": 2.182, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.0125839710235596, | |
| "eval_runtime": 144.1418, | |
| "eval_samples_per_second": 43.527, | |
| "eval_steps_per_second": 5.446, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9626, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 1.9195271730422974, | |
| "eval_runtime": 144.2253, | |
| "eval_samples_per_second": 43.501, | |
| "eval_steps_per_second": 5.443, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8881, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 1.858821988105774, | |
| "eval_runtime": 144.0313, | |
| "eval_samples_per_second": 43.56, | |
| "eval_steps_per_second": 5.45, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8367, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.8125503063201904, | |
| "eval_runtime": 143.9861, | |
| "eval_samples_per_second": 43.574, | |
| "eval_steps_per_second": 5.452, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 1.785, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.7720445394515991, | |
| "eval_runtime": 144.0061, | |
| "eval_samples_per_second": 43.568, | |
| "eval_steps_per_second": 5.451, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7562, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.7424511909484863, | |
| "eval_runtime": 144.0245, | |
| "eval_samples_per_second": 43.562, | |
| "eval_steps_per_second": 5.45, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7176, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 1.7112771272659302, | |
| "eval_runtime": 144.2204, | |
| "eval_samples_per_second": 43.503, | |
| "eval_steps_per_second": 5.443, | |
| "step": 700 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 4.174060032521011e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |