| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 22944, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.996078431372549e-05, | |
| "loss": 1.0033, | |
| "step": 1529 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.8894377451692577e-05, | |
| "loss": 0.8149, | |
| "step": 3058 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.667296237105913e-05, | |
| "loss": 0.7863, | |
| "step": 4587 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2.4453000145285484e-05, | |
| "loss": 0.7733, | |
| "step": 6116 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.223158506465204e-05, | |
| "loss": 0.7488, | |
| "step": 7645 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 2.00116228388784e-05, | |
| "loss": 0.7485, | |
| "step": 9174 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.779020775824495e-05, | |
| "loss": 0.7375, | |
| "step": 10703 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.5570245532471306e-05, | |
| "loss": 0.7229, | |
| "step": 12232 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.3348830451837863e-05, | |
| "loss": 0.721, | |
| "step": 13761 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.1127415371204417e-05, | |
| "loss": 0.7253, | |
| "step": 15290 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 8.907453145430772e-06, | |
| "loss": 0.7086, | |
| "step": 16819 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 6.688943774516926e-06, | |
| "loss": 0.7119, | |
| "step": 18348 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.467528693883481e-06, | |
| "loss": 0.7104, | |
| "step": 19877 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 2.246113613250036e-06, | |
| "loss": 0.7096, | |
| "step": 21406 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 2.7604242336190614e-08, | |
| "loss": 0.7066, | |
| "step": 22935 | |
| } | |
| ], | |
| "logging_steps": 1529, | |
| "max_steps": 22944, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.8652122334919393e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |