| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 246, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0002, | |
| "loss": 3.4918, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0002, | |
| "loss": 3.4392, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0002, | |
| "loss": 3.3662, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0002, | |
| "loss": 3.23, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0002, | |
| "loss": 3.0296, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0002, | |
| "loss": 2.9953, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0002, | |
| "loss": 2.9222, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0002, | |
| "loss": 2.7419, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6997, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6147, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6041, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5295, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.1770684719085693, | |
| "eval_runtime": 5.1953, | |
| "eval_samples_per_second": 2.887, | |
| "eval_steps_per_second": 0.385, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5208, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4462, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4549, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5357, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4114, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4352, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4544, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.0002, | |
| "loss": 2.399, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4452, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3952, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4314, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.0628561973571777, | |
| "eval_runtime": 5.0293, | |
| "eval_samples_per_second": 2.983, | |
| "eval_steps_per_second": 0.398, | |
| "step": 246 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 615, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 2617504903987200.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |