| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 155, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 3.5674822330474854, | |
| "learning_rate": 0.0001375, | |
| "loss": 3.066, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 2.487783670425415, | |
| "learning_rate": 0.00019136690647482017, | |
| "loss": 2.2172, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 3.1076672077178955, | |
| "learning_rate": 0.00017553956834532374, | |
| "loss": 1.4606, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 2.9141297340393066, | |
| "learning_rate": 0.00015971223021582736, | |
| "loss": 1.1038, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 2.831645965576172, | |
| "learning_rate": 0.00014388489208633093, | |
| "loss": 1.0646, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 2.672914743423462, | |
| "learning_rate": 0.00012805755395683453, | |
| "loss": 1.0186, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 2.9274446964263916, | |
| "learning_rate": 0.00011223021582733813, | |
| "loss": 0.8428, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 3.164165496826172, | |
| "learning_rate": 9.640287769784174e-05, | |
| "loss": 0.7633, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "grad_norm": 4.559382438659668, | |
| "learning_rate": 8.057553956834533e-05, | |
| "loss": 0.6851, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "grad_norm": 2.335391044616699, | |
| "learning_rate": 6.474820143884892e-05, | |
| "loss": 0.6521, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "grad_norm": 3.2239022254943848, | |
| "learning_rate": 4.892086330935252e-05, | |
| "loss": 0.706, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "grad_norm": 2.8791332244873047, | |
| "learning_rate": 3.3093525179856116e-05, | |
| "loss": 0.6047, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "grad_norm": 3.5159878730773926, | |
| "learning_rate": 1.7266187050359716e-05, | |
| "loss": 0.4889, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "grad_norm": 3.642343044281006, | |
| "learning_rate": 1.4388489208633094e-06, | |
| "loss": 0.585, | |
| "step": 154 | |
| } | |
| ], | |
| "logging_steps": 11, | |
| "max_steps": 155, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 6751637581824000.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |