| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 462, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 4.875, | |
| "learning_rate": 0.0002, | |
| "loss": 18.0666, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 8.625, | |
| "learning_rate": 0.0002, | |
| "loss": 11.9963, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 3.84375, | |
| "learning_rate": 0.0002, | |
| "loss": 6.0667, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 2.375, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4769, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 2.109375, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2233, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 1.8984375, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1739, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 3.375, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0811, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 4.0, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9928, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 1.4140625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9649, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 1.7421875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9645, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 2.171875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9812, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 1.5546875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9106, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 2.984375, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8943, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 1.3046875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9205, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 2.234375, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9445, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8882, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 2.984375, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8806, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 2.46875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8692, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 1.4375, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8291, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 1.515625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8069, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 2.671875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8406, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8694, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.8671875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.821, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 2.03125, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8243, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7678, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 2.0, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8301, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8845, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 1.078125, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8901, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 1.015625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8198, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7615, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 1.0703125, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8717, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 1.515625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7862, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 1.6796875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8674, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 1.0859375, | |
| "learning_rate": 0.0002, | |
| "loss": 1.822, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.921875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7834, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 1.4921875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8157, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7164, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.835, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 1.2109375, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7927, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 1.8046875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7038, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7914, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 2.921875, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7892, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.83984375, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7335, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 1.453125, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8013, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 1.5078125, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7118, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.4140625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8202, | |
| "step": 460 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1386, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.302644915577815e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |