| { | |
| "best_metric": 0.47078712112764887, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-4/checkpoint-3207", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 3207, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 4.060019016265869, | |
| "learning_rate": 4.481496253268653e-05, | |
| "loss": 0.5778, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 12.268152236938477, | |
| "learning_rate": 3.6537355858751076e-05, | |
| "loss": 0.5395, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.4984060525894165, | |
| "eval_matthews_correlation": 0.39596800444544106, | |
| "eval_runtime": 0.6991, | |
| "eval_samples_per_second": 1491.822, | |
| "eval_steps_per_second": 94.401, | |
| "step": 1069 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 4.944720268249512, | |
| "learning_rate": 2.8259749184815627e-05, | |
| "loss": 0.3959, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 9.598797798156738, | |
| "learning_rate": 1.9982142510880178e-05, | |
| "loss": 0.3603, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5668240785598755, | |
| "eval_matthews_correlation": 0.43881557658403864, | |
| "eval_runtime": 1.2379, | |
| "eval_samples_per_second": 842.559, | |
| "eval_steps_per_second": 53.316, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 0.8520996570587158, | |
| "learning_rate": 1.1704535836944727e-05, | |
| "loss": 0.2352, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 1.3932373523712158, | |
| "learning_rate": 3.426929163009277e-06, | |
| "loss": 0.2377, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.8578255772590637, | |
| "eval_matthews_correlation": 0.47078712112764887, | |
| "eval_runtime": 1.3062, | |
| "eval_samples_per_second": 798.53, | |
| "eval_steps_per_second": 50.53, | |
| "step": 3207 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3207, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 112923282846480.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 5.309256920662198e-05, | |
| "num_train_epochs": 3, | |
| "per_device_train_batch_size": 8, | |
| "seed": 37 | |
| } | |
| } | |