| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 150.0, | |
| "eval_steps": 0, | |
| "global_step": 150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 28.858620901314232, | |
| "learning_rate": 9.59731543624161e-06, | |
| "loss": -0.5725, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 40.552036143593824, | |
| "learning_rate": 9.12751677852349e-06, | |
| "loss": -0.0209, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "grad_norm": 41.64540321542807, | |
| "learning_rate": 8.65771812080537e-06, | |
| "loss": -1.6992, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 49.85638001833566, | |
| "learning_rate": 8.18791946308725e-06, | |
| "loss": -1.5316, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 51.51197916884997, | |
| "learning_rate": 7.718120805369127e-06, | |
| "loss": -2.5143, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "grad_norm": 48.86359832099742, | |
| "learning_rate": 7.248322147651007e-06, | |
| "loss": -4.165, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "grad_norm": 51.03975313916726, | |
| "learning_rate": 6.778523489932887e-06, | |
| "loss": -5.3398, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "grad_norm": 52.33035721537868, | |
| "learning_rate": 6.308724832214766e-06, | |
| "loss": -6.3177, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "grad_norm": 58.038819315232836, | |
| "learning_rate": 5.8389261744966455e-06, | |
| "loss": -8.2965, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 67.83918432181187, | |
| "learning_rate": 5.369127516778524e-06, | |
| "loss": -8.528, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "grad_norm": 79.13375379819861, | |
| "learning_rate": 4.899328859060403e-06, | |
| "loss": -10.5042, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "grad_norm": 81.13553186716261, | |
| "learning_rate": 4.429530201342283e-06, | |
| "loss": -10.9264, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "grad_norm": 85.2509810694198, | |
| "learning_rate": 3.959731543624161e-06, | |
| "loss": -12.0697, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "grad_norm": 78.38132339540856, | |
| "learning_rate": 3.4899328859060407e-06, | |
| "loss": -13.1855, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 105.0, | |
| "grad_norm": 84.10584475490751, | |
| "learning_rate": 3.02013422818792e-06, | |
| "loss": -15.2029, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 112.0, | |
| "grad_norm": 83.51387820229583, | |
| "learning_rate": 2.5503355704697992e-06, | |
| "loss": -16.0523, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 119.0, | |
| "grad_norm": 84.75322452499404, | |
| "learning_rate": 2.080536912751678e-06, | |
| "loss": -16.7421, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 126.0, | |
| "grad_norm": 85.1842427450598, | |
| "learning_rate": 1.6107382550335572e-06, | |
| "loss": -17.246, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 133.0, | |
| "grad_norm": 93.8841349140048, | |
| "learning_rate": 1.1409395973154363e-06, | |
| "loss": -17.8875, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 140.0, | |
| "grad_norm": 88.93428899110127, | |
| "learning_rate": 6.711409395973155e-07, | |
| "loss": -18.4957, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 147.0, | |
| "grad_norm": 89.46815374400903, | |
| "learning_rate": 2.0134228187919465e-07, | |
| "loss": -18.9681, | |
| "step": 147 | |
| } | |
| ], | |
| "logging_steps": 7, | |
| "max_steps": 150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 150, | |
| "save_steps": 15, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |