| { | |
| "best_metric": 0.514713087977408, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-1/checkpoint-2140", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 2140, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.894237945372658e-05, | |
| "loss": 0.5408, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.48756271600723267, | |
| "eval_matthews_correlation": 0.4371151083758757, | |
| "eval_runtime": 0.7652, | |
| "eval_samples_per_second": 1363.003, | |
| "eval_steps_per_second": 86.249, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.769125774022621e-05, | |
| "loss": 0.3186, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.4699440896511078, | |
| "eval_matthews_correlation": 0.48170038113970914, | |
| "eval_runtime": 0.7508, | |
| "eval_samples_per_second": 1389.228, | |
| "eval_steps_per_second": 87.909, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 2.644013602672585e-05, | |
| "loss": 0.2026, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.7681801319122314, | |
| "eval_matthews_correlation": 0.47428224772244454, | |
| "eval_runtime": 0.7372, | |
| "eval_samples_per_second": 1414.877, | |
| "eval_steps_per_second": 89.532, | |
| "step": 1605 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 1.5189014313225487e-05, | |
| "loss": 0.1282, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8975861072540283, | |
| "eval_matthews_correlation": 0.514713087977408, | |
| "eval_runtime": 1.2396, | |
| "eval_samples_per_second": 841.372, | |
| "eval_steps_per_second": 53.241, | |
| "step": 2140 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2675, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 171168835819980.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 6.019350116722694e-05, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 16, | |
| "seed": 13 | |
| } | |
| } | |