| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.9384615384615387, |
| "eval_steps": 500, |
| "global_step": 208, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.19, |
| "grad_norm": 0.2056211680173874, |
| "learning_rate": 0.0002, |
| "loss": 1.209, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.38, |
| "grad_norm": 0.20694544911384583, |
| "learning_rate": 0.0002, |
| "loss": 1.1006, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.57, |
| "grad_norm": 0.08869898319244385, |
| "learning_rate": 0.0002, |
| "loss": 0.9996, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.76, |
| "grad_norm": 0.1591188907623291, |
| "learning_rate": 0.0002, |
| "loss": 0.9352, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.95, |
| "grad_norm": 0.21296942234039307, |
| "learning_rate": 0.0002, |
| "loss": 0.9077, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.98, |
| "eval_loss": 1.0426653623580933, |
| "eval_runtime": 5.957, |
| "eval_samples_per_second": 16.787, |
| "eval_steps_per_second": 0.336, |
| "step": 52 |
| }, |
| { |
| "epoch": 1.14, |
| "grad_norm": 0.09350233525037766, |
| "learning_rate": 0.0002, |
| "loss": 1.0513, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.33, |
| "grad_norm": 0.10060329735279083, |
| "learning_rate": 0.0002, |
| "loss": 0.8996, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.51, |
| "grad_norm": 0.08181482553482056, |
| "learning_rate": 0.0002, |
| "loss": 0.839, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.7, |
| "grad_norm": 0.1657375991344452, |
| "learning_rate": 0.0002, |
| "loss": 0.8408, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.89, |
| "grad_norm": 0.1462504267692566, |
| "learning_rate": 0.0002, |
| "loss": 0.9489, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.99, |
| "eval_loss": 1.0188666582107544, |
| "eval_runtime": 5.5117, |
| "eval_samples_per_second": 18.143, |
| "eval_steps_per_second": 0.363, |
| "step": 105 |
| }, |
| { |
| "epoch": 2.08, |
| "grad_norm": 0.06421244889497757, |
| "learning_rate": 0.0002, |
| "loss": 0.9165, |
| "step": 110 |
| }, |
| { |
| "epoch": 2.27, |
| "grad_norm": 0.19528228044509888, |
| "learning_rate": 0.0002, |
| "loss": 0.7311, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.46, |
| "grad_norm": 0.1915639042854309, |
| "learning_rate": 0.0002, |
| "loss": 0.7609, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.65, |
| "grad_norm": 0.14902861416339874, |
| "learning_rate": 0.0002, |
| "loss": 0.8944, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.84, |
| "grad_norm": 0.11538069695234299, |
| "learning_rate": 0.0002, |
| "loss": 0.8738, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.99, |
| "eval_loss": 1.0385792255401611, |
| "eval_runtime": 5.6367, |
| "eval_samples_per_second": 17.741, |
| "eval_steps_per_second": 0.355, |
| "step": 158 |
| }, |
| { |
| "epoch": 3.03, |
| "grad_norm": 0.07593070715665817, |
| "learning_rate": 0.0002, |
| "loss": 0.6973, |
| "step": 160 |
| }, |
| { |
| "epoch": 3.22, |
| "grad_norm": 0.18991707265377045, |
| "learning_rate": 0.0002, |
| "loss": 0.6857, |
| "step": 170 |
| }, |
| { |
| "epoch": 3.41, |
| "grad_norm": 0.16755373775959015, |
| "learning_rate": 0.0002, |
| "loss": 0.8125, |
| "step": 180 |
| }, |
| { |
| "epoch": 3.6, |
| "grad_norm": 0.11783052235841751, |
| "learning_rate": 0.0002, |
| "loss": 0.7726, |
| "step": 190 |
| }, |
| { |
| "epoch": 3.79, |
| "grad_norm": 0.11990819126367569, |
| "learning_rate": 0.0002, |
| "loss": 0.6837, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.94, |
| "eval_loss": 1.0604676008224487, |
| "eval_runtime": 1.1256, |
| "eval_samples_per_second": 88.843, |
| "eval_steps_per_second": 1.777, |
| "step": 208 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 208, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "save_steps": 250, |
| "total_flos": 1.4978204330287432e+18, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
