| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9916434540389973, | |
| "eval_steps": 500, | |
| "global_step": 537, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2e-05, | |
| "loss": 95.476, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2e-05, | |
| "loss": 48.4218, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 2e-05, | |
| "loss": 13.7763, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 2e-05, | |
| "loss": 1.648, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4463, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9103, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6693, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8188, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7446, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6158, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9035, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9486, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7198, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 2e-05, | |
| "loss": 1.5337, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 2e-05, | |
| "loss": 1.3312, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8632, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8279, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9823, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7963, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7861, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9859, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0639, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7174, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7996, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7602, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8271, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "step": 537, | |
| "total_flos": 1345715322224640.0, | |
| "train_loss": 6.7171177713120676, | |
| "train_runtime": 1535.649, | |
| "train_samples_per_second": 11.219, | |
| "train_steps_per_second": 0.35 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 537, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1345715322224640.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |