| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 684, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 1.8538011695906434e-05, | |
| "loss": 2.6775, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.7076023391812867e-05, | |
| "loss": 2.4413, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.56140350877193e-05, | |
| "loss": 1.9558, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.4152046783625733e-05, | |
| "loss": 1.5483, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.2690058479532166e-05, | |
| "loss": 1.3034, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 1.1228070175438597e-05, | |
| "loss": 1.0546, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 9.76608187134503e-06, | |
| "loss": 0.9354, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 8.304093567251463e-06, | |
| "loss": 0.9152, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 6.842105263157896e-06, | |
| "loss": 0.8912, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 5.380116959064328e-06, | |
| "loss": 0.7445, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_accuracy": 0.7962861824139814, | |
| "eval_loss": 0.8156927824020386, | |
| "eval_runtime": 4.3002, | |
| "eval_samples_per_second": 425.794, | |
| "eval_steps_per_second": 26.743, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 3.9181286549707605e-06, | |
| "loss": 0.6671, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 2.456140350877193e-06, | |
| "loss": 0.7009, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 9.941520467836258e-07, | |
| "loss": 0.6977, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 684, | |
| "total_flos": 995075069121756.0, | |
| "train_loss": 1.2412931403221443, | |
| "train_runtime": 161.3672, | |
| "train_samples_per_second": 135.585, | |
| "train_steps_per_second": 4.239 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 684, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 995075069121756.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |