| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.968, | |
| "eval_steps": 62, | |
| "global_step": 248, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 1.5051484779217006, | |
| "learning_rate": 1.2096774193548388e-05, | |
| "loss": 0.0697, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.6557213411275504, | |
| "learning_rate": 2.4193548387096777e-05, | |
| "loss": 0.0453, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 5.8190678296309155, | |
| "learning_rate": 3.6290322580645165e-05, | |
| "loss": 0.1379, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 5.082131701290993, | |
| "learning_rate": 4.8387096774193554e-05, | |
| "loss": 0.3262, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 4.533397429277955, | |
| "learning_rate": 4.74e-05, | |
| "loss": 0.4413, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 3.9912559897316915, | |
| "learning_rate": 4.44e-05, | |
| "loss": 0.4631, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 3.605033658976913, | |
| "learning_rate": 4.14e-05, | |
| "loss": 0.3972, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 3.1324872605719025, | |
| "learning_rate": 3.8400000000000005e-05, | |
| "loss": 0.3926, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 2.7954439766075647, | |
| "learning_rate": 3.54e-05, | |
| "loss": 0.3268, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 2.4279604222176263, | |
| "learning_rate": 3.24e-05, | |
| "loss": 0.2419, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 2.2595799598751447, | |
| "learning_rate": 2.94e-05, | |
| "loss": 0.2212, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 2.573622729528534, | |
| "learning_rate": 2.64e-05, | |
| "loss": 0.1586, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 2.2604733828967016, | |
| "learning_rate": 2.3400000000000003e-05, | |
| "loss": 0.1431, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 1.4987841874421017, | |
| "learning_rate": 2.04e-05, | |
| "loss": 0.1047, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.6724445238800598, | |
| "learning_rate": 1.74e-05, | |
| "loss": 0.0896, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 1.3552689605969652, | |
| "learning_rate": 1.44e-05, | |
| "loss": 0.0871, | |
| "step": 240 | |
| } | |
| ], | |
| "logging_steps": 15, | |
| "max_steps": 312, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 62, | |
| "total_flos": 0.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |