| { | |
| "best_metric": 0.18187156319618225, | |
| "best_model_checkpoint": "mgh6/TCS_Pair/run-bad69_00000/checkpoint-2000", | |
| "epoch": 0.9412397472098964, | |
| "eval_steps": 500, | |
| "global_step": 3500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00079158678443389, | |
| "loss": 0.1584, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 0.2451040893793106, | |
| "eval_runtime": 21.3792, | |
| "eval_samples_per_second": 224.096, | |
| "eval_steps_per_second": 28.018, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.000786228087003387, | |
| "loss": 0.1285, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 0.19461262226104736, | |
| "eval_runtime": 21.0052, | |
| "eval_samples_per_second": 228.086, | |
| "eval_steps_per_second": 28.517, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0007808693895728842, | |
| "loss": 0.1232, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.268852174282074, | |
| "eval_runtime": 21.1069, | |
| "eval_samples_per_second": 226.988, | |
| "eval_steps_per_second": 28.379, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0007755106921423813, | |
| "loss": 0.1182, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_loss": 0.18187156319618225, | |
| "eval_runtime": 21.0753, | |
| "eval_samples_per_second": 227.328, | |
| "eval_steps_per_second": 28.422, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0007701519947118784, | |
| "loss": 0.1642, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 0.25392478704452515, | |
| "eval_runtime": 21.48, | |
| "eval_samples_per_second": 223.045, | |
| "eval_steps_per_second": 27.886, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.0007647932972813753, | |
| "loss": 0.1779, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_loss": 0.2446264624595642, | |
| "eval_runtime": 21.3207, | |
| "eval_samples_per_second": 224.711, | |
| "eval_steps_per_second": 28.095, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.0007594345998508726, | |
| "loss": 0.1776, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_loss": 0.2562008798122406, | |
| "eval_runtime": 21.0095, | |
| "eval_samples_per_second": 228.039, | |
| "eval_steps_per_second": 28.511, | |
| "step": 3500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 74360, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 3555307819392000.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "classifier_dropout": 0.18727005942368125, | |
| "gradient_accumulation_steps": 2, | |
| "learning_rate": 0.0007969454818643929, | |
| "weight_decay": 0.029106359131330688 | |
| } | |
| } | |