| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 25.0, | |
| "eval_steps": 200, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.125, | |
| "eval_loss": 2.9946672916412354, | |
| "eval_runtime": 5.136, | |
| "eval_samples_per_second": 292.251, | |
| "eval_steps_per_second": 3.31, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 0.0002, | |
| "loss": 2.574, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.41015625, | |
| "learning_rate": 0.00019863613034027224, | |
| "loss": 2.3858, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.22265625, | |
| "learning_rate": 0.00019458172417006347, | |
| "loss": 2.2927, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.177734375, | |
| "learning_rate": 0.0001879473751206489, | |
| "loss": 2.2352, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 0.1513671875, | |
| "learning_rate": 0.00017891405093963938, | |
| "loss": 2.2042, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.181640625, | |
| "learning_rate": 0.00016772815716257412, | |
| "loss": 2.1814, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 0.1240234375, | |
| "learning_rate": 0.00015469481581224272, | |
| "loss": 2.1483, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.1494140625, | |
| "learning_rate": 0.00014016954246529696, | |
| "loss": 2.1455, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "grad_norm": 0.125, | |
| "learning_rate": 0.00012454854871407994, | |
| "loss": 2.1361, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 0.201171875, | |
| "learning_rate": 0.00010825793454723325, | |
| "loss": 2.109, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "grad_norm": 0.1259765625, | |
| "learning_rate": 9.174206545276677e-05, | |
| "loss": 2.106, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.1279296875, | |
| "learning_rate": 7.54514512859201e-05, | |
| "loss": 2.0939, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "grad_norm": 0.13671875, | |
| "learning_rate": 5.983045753470308e-05, | |
| "loss": 2.0901, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "grad_norm": 0.1396484375, | |
| "learning_rate": 4.530518418775733e-05, | |
| "loss": 2.0959, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "grad_norm": 0.138671875, | |
| "learning_rate": 3.227184283742591e-05, | |
| "loss": 2.0851, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.1259765625, | |
| "learning_rate": 2.1085949060360654e-05, | |
| "loss": 2.0874, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "grad_norm": 0.1240234375, | |
| "learning_rate": 1.2052624879351104e-05, | |
| "loss": 2.0865, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "grad_norm": 0.16796875, | |
| "learning_rate": 5.418275829936537e-06, | |
| "loss": 2.0931, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 23.75, | |
| "grad_norm": 0.1435546875, | |
| "learning_rate": 1.3638696597277679e-06, | |
| "loss": 2.0826, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 0.142578125, | |
| "learning_rate": 0.0, | |
| "loss": 2.0872, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 2.159134864807129, | |
| "eval_runtime": 5.0049, | |
| "eval_samples_per_second": 299.904, | |
| "eval_steps_per_second": 3.397, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 25, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.113876738932736e+16, | |
| "train_batch_size": 23, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |