| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2708, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 1.9941379571543597e-05, | |
| "loss": 0.2708, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1.976620555710087e-05, | |
| "loss": 0.4364, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 1.9476531711828027e-05, | |
| "loss": 0.2522, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.9075754196709574e-05, | |
| "loss": 0.1957, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 1.8568571761675893e-05, | |
| "loss": 0.1807, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.796093065705644e-05, | |
| "loss": 0.2076, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.725995491923131e-05, | |
| "loss": 0.196, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.647386284781828e-05, | |
| "loss": 0.1909, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.5611870653623826e-05, | |
| "loss": 0.1874, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 1.4684084406997903e-05, | |
| "loss": 0.1771, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.3701381553399147e-05, | |
| "loss": 0.1791, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 1.2675283385292212e-05, | |
| "loss": 0.172, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 1.161781996552765e-05, | |
| "loss": 0.1626, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.0541389085854177e-05, | |
| "loss": 0.16, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 9.458610914145826e-06, | |
| "loss": 0.1503, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 8.382180034472353e-06, | |
| "loss": 0.138, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 7.324716614707794e-06, | |
| "loss": 0.1295, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 6.298618446600856e-06, | |
| "loss": 0.1332, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 5.3159155930021e-06, | |
| "loss": 0.1234, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 4.388129346376177e-06, | |
| "loss": 0.1118, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 3.5261371521817247e-06, | |
| "loss": 0.1146, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 2.740045080768694e-06, | |
| "loss": 0.119, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 2.0390693429435626e-06, | |
| "loss": 0.1026, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.4314282383241097e-06, | |
| "loss": 0.108, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 9.242458032904311e-07, | |
| "loss": 0.0969, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "learning_rate": 5.234682881719766e-07, | |
| "loss": 0.1, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 2.3379444289913344e-07, | |
| "loss": 0.09, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 5.862042845640403e-08, | |
| "loss": 0.0873, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0986, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 30, | |
| "total_flos": 711634452480.0, | |
| "train_loss": 0.16474711100260417, | |
| "train_runtime": 357.5156, | |
| "train_samples_per_second": 5.337, | |
| "train_steps_per_second": 0.084 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 111, | |
| "total_flos": 711634452480.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |