| { | |
| "best_metric": 0.40039137901723904, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola\\run-0\\checkpoint-5345", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 5345, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 3.136430025100708, | |
| "learning_rate": 1.057670968337005e-06, | |
| "loss": 0.6141, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 3.4976818561553955, | |
| "learning_rate": 9.485201976107919e-07, | |
| "loss": 0.5835, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5739973187446594, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.1959, | |
| "eval_samples_per_second": 474.985, | |
| "eval_steps_per_second": 30.057, | |
| "step": 1069 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 5.231539249420166, | |
| "learning_rate": 8.393694268845788e-07, | |
| "loss": 0.5608, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 7.4941792488098145, | |
| "learning_rate": 7.302186561583657e-07, | |
| "loss": 0.5232, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5443820953369141, | |
| "eval_matthews_correlation": 0.2045620264226157, | |
| "eval_runtime": 0.9649, | |
| "eval_samples_per_second": 1080.926, | |
| "eval_steps_per_second": 68.4, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 7.5233941078186035, | |
| "learning_rate": 6.210678854321524e-07, | |
| "loss": 0.5208, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 6.596949100494385, | |
| "learning_rate": 5.119171147059394e-07, | |
| "loss": 0.4937, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.5387976169586182, | |
| "eval_matthews_correlation": 0.3459293076636948, | |
| "eval_runtime": 0.8998, | |
| "eval_samples_per_second": 1159.164, | |
| "eval_steps_per_second": 73.351, | |
| "step": 3207 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "grad_norm": 6.339045524597168, | |
| "learning_rate": 4.0276634397972634e-07, | |
| "loss": 0.4854, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "grad_norm": 12.96947193145752, | |
| "learning_rate": 2.936155732535133e-07, | |
| "loss": 0.4801, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.5404464602470398, | |
| "eval_matthews_correlation": 0.3860262930745603, | |
| "eval_runtime": 0.852, | |
| "eval_samples_per_second": 1224.144, | |
| "eval_steps_per_second": 77.463, | |
| "step": 4276 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "grad_norm": 15.995396614074707, | |
| "learning_rate": 1.8446480252730014e-07, | |
| "loss": 0.4664, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "grad_norm": 14.75170612335205, | |
| "learning_rate": 7.531403180108705e-08, | |
| "loss": 0.4841, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.5346407890319824, | |
| "eval_matthews_correlation": 0.40039137901723904, | |
| "eval_runtime": 1.0251, | |
| "eval_samples_per_second": 1017.419, | |
| "eval_steps_per_second": 64.381, | |
| "step": 5345 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5345, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 188608220598120.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.166821739063218e-06, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 8, | |
| "seed": 23 | |
| } | |
| } | |