| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.860775588048315, | |
| "eval_steps": 500, | |
| "global_step": 9000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 18.022613525390625, | |
| "learning_rate": 4.735113371477008e-05, | |
| "loss": 2.3719, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 21.23277473449707, | |
| "learning_rate": 4.4702267429540156e-05, | |
| "loss": 2.0069, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 16.839088439941406, | |
| "learning_rate": 4.205340114431024e-05, | |
| "loss": 1.8155, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 15.446898460388184, | |
| "learning_rate": 3.9404534859080315e-05, | |
| "loss": 1.7476, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 24.382474899291992, | |
| "learning_rate": 3.675566857385039e-05, | |
| "loss": 1.6636, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 21.278900146484375, | |
| "learning_rate": 3.410680228862047e-05, | |
| "loss": 1.624, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 14.305024147033691, | |
| "learning_rate": 3.1457936003390545e-05, | |
| "loss": 1.2214, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 13.627291679382324, | |
| "learning_rate": 2.8809069718160632e-05, | |
| "loss": 1.0202, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 11.309160232543945, | |
| "learning_rate": 2.616020343293071e-05, | |
| "loss": 1.0545, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 12.876075744628906, | |
| "learning_rate": 2.3511337147700785e-05, | |
| "loss": 1.0358, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 11.487010955810547, | |
| "learning_rate": 2.0862470862470865e-05, | |
| "loss": 1.0278, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 12.434341430664062, | |
| "learning_rate": 1.821360457724094e-05, | |
| "loss": 1.0069, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 11.067166328430176, | |
| "learning_rate": 1.5564738292011018e-05, | |
| "loss": 0.8587, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "grad_norm": 9.134782791137695, | |
| "learning_rate": 1.29158720067811e-05, | |
| "loss": 0.6709, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "grad_norm": 12.168549537658691, | |
| "learning_rate": 1.0267005721551176e-05, | |
| "loss": 0.6534, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 8.282147407531738, | |
| "learning_rate": 7.6181394363212545e-06, | |
| "loss": 0.6821, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 22.135112762451172, | |
| "learning_rate": 4.9692731510913336e-06, | |
| "loss": 0.671, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "grad_norm": 7.933387756347656, | |
| "learning_rate": 2.3204068658614114e-06, | |
| "loss": 0.669, | |
| "step": 9000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 9438, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 184716162957312.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |