| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "eval_steps": 500, | |
| "global_step": 4700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 10.64, | |
| "learning_rate": 4.468085106382979e-05, | |
| "loss": 0.77, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 21.28, | |
| "learning_rate": 3.936170212765958e-05, | |
| "loss": 0.0564, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 31.91, | |
| "learning_rate": 3.4042553191489365e-05, | |
| "loss": 0.0198, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 42.55, | |
| "learning_rate": 2.8723404255319154e-05, | |
| "loss": 0.0113, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 53.19, | |
| "learning_rate": 2.340425531914894e-05, | |
| "loss": 0.0061, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 63.83, | |
| "learning_rate": 1.8085106382978724e-05, | |
| "loss": 0.0063, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 74.47, | |
| "learning_rate": 1.2765957446808511e-05, | |
| "loss": 0.0077, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 85.11, | |
| "learning_rate": 7.446808510638298e-06, | |
| "loss": 0.0113, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 95.74, | |
| "learning_rate": 2.1276595744680853e-06, | |
| "loss": 0.023, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 4700, | |
| "total_flos": 5.87917702656e+16, | |
| "train_loss": 0.09985893122693326, | |
| "train_runtime": 2075.477, | |
| "train_samples_per_second": 144.545, | |
| "train_steps_per_second": 2.265 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4700, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "total_flos": 5.87917702656e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |