| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.2248995983935744, | |
| "global_step": 69, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0008709677419354839, | |
| "loss": 2.7331, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0007419354838709678, | |
| "loss": 2.6157, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.0006129032258064516, | |
| "loss": 2.566, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0004838709677419355, | |
| "loss": 2.48, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0003548387096774194, | |
| "loss": 2.3692, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00022580645161290321, | |
| "loss": 2.2791, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.0006989247311827958, | |
| "loss": 2.2783, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 20.55, | |
| "eval_loss": 2.021012306213379, | |
| "eval_rouge1": 32.3793, | |
| "eval_rouge2": 12.6427, | |
| "eval_rougeL": 27.4027, | |
| "eval_rougeLsum": 27.3, | |
| "eval_runtime": 3629.2928, | |
| "eval_samples_per_second": 0.028, | |
| "eval_steps_per_second": 0.009, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.0006559139784946236, | |
| "loss": 2.2677, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.0006129032258064516, | |
| "loss": 1.8913, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.0005698924731182796, | |
| "loss": 1.9234, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.0005268817204301075, | |
| "loss": 1.8524, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.0004838709677419355, | |
| "loss": 1.8549, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00044086021505376343, | |
| "loss": 1.9571, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.0003978494623655914, | |
| "loss": 1.8606, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.0003548387096774194, | |
| "loss": 1.791, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 21.6, | |
| "eval_loss": 1.9703446626663208, | |
| "eval_rouge1": 34.5495, | |
| "eval_rouge2": 12.6789, | |
| "eval_rougeL": 28.3741, | |
| "eval_rougeLsum": 28.4386, | |
| "eval_runtime": 3780.972, | |
| "eval_samples_per_second": 0.026, | |
| "eval_steps_per_second": 0.009, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 0.0003118279569892473, | |
| "loss": 1.7687, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.00026881720430107527, | |
| "loss": 1.5525, | |
| "step": 68 | |
| } | |
| ], | |
| "max_steps": 93, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.813836532618199e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |