| { | |
| "best_metric": 0.00037940836045891047, | |
| "best_model_checkpoint": "data/further_pretraining/bert-base-uncased/checkpoint-2155", | |
| "epoch": 2.0, | |
| "global_step": 4310, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.9535962877030165e-05, | |
| "loss": 0.0545, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1.9071925754060325e-05, | |
| "loss": 0.0012, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.860788863109049e-05, | |
| "loss": 0.0017, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1.8143851508120652e-05, | |
| "loss": 0.0003, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.00037940836045891047, | |
| "eval_runtime": 164.4067, | |
| "eval_samples_per_second": 419.32, | |
| "eval_steps_per_second": 13.108, | |
| "step": 2155 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 1.7679814385150812e-05, | |
| "loss": 0.0006, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.7215777262180976e-05, | |
| "loss": 0.0005, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 1.675174013921114e-05, | |
| "loss": 0.001, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.62877030162413e-05, | |
| "loss": 0.0007, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.00047821112093515694, | |
| "eval_runtime": 161.3062, | |
| "eval_samples_per_second": 427.38, | |
| "eval_steps_per_second": 13.36, | |
| "step": 4310 | |
| } | |
| ], | |
| "max_steps": 21550, | |
| "num_train_epochs": 10, | |
| "total_flos": 9072538294809600.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |