| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 1266, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.9606299212598425e-05, | |
| "loss": 4.3298, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.8946444249341527e-05, | |
| "loss": 4.1173, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.6733977172958736e-05, | |
| "loss": 3.8942, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2.4521510096575942e-05, | |
| "loss": 3.8492, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.2309043020193154e-05, | |
| "loss": 3.7421, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 2.009657594381036e-05, | |
| "loss": 3.71, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.788410886742757e-05, | |
| "loss": 3.639, | |
| "step": 588 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 1.5671641791044774e-05, | |
| "loss": 3.6233, | |
| "step": 672 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 1.3459174714661984e-05, | |
| "loss": 3.6277, | |
| "step": 756 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.1246707638279193e-05, | |
| "loss": 3.5984, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 9.0342405618964e-06, | |
| "loss": 3.5838, | |
| "step": 924 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 6.8217734855136085e-06, | |
| "loss": 3.5981, | |
| "step": 1008 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 4.6093064091308165e-06, | |
| "loss": 3.5963, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 2.396839332748025e-06, | |
| "loss": 3.5397, | |
| "step": 1176 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 1.8437225636523267e-07, | |
| "loss": 3.5674, | |
| "step": 1260 | |
| } | |
| ], | |
| "logging_steps": 84, | |
| "max_steps": 1266, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 4721807550578688.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |