| { | |
| "best_metric": 0.4605477899795795, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-1/checkpoint-4276", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 4276, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 21.092390060424805, | |
| "learning_rate": 1.2650690583673927e-05, | |
| "loss": 0.5742, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 14.845769882202148, | |
| "learning_rate": 1.0975546173759478e-05, | |
| "loss": 0.5397, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 13.874490737915039, | |
| "learning_rate": 9.300401763845028e-06, | |
| "loss": 0.5327, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 22.50566864013672, | |
| "learning_rate": 7.625257353930576e-06, | |
| "loss": 0.5138, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6036589741706848, | |
| "eval_matthews_correlation": 0.40976281740188225, | |
| "eval_runtime": 0.3434, | |
| "eval_samples_per_second": 3036.988, | |
| "eval_steps_per_second": 192.178, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 0.8492972254753113, | |
| "learning_rate": 5.950112944016127e-06, | |
| "loss": 0.4239, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 1.9886528253555298, | |
| "learning_rate": 4.274968534101676e-06, | |
| "loss": 0.4516, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 143.33644104003906, | |
| "learning_rate": 2.5998241241872264e-06, | |
| "loss": 0.4565, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 96.76008605957031, | |
| "learning_rate": 9.246797142727765e-07, | |
| "loss": 0.4628, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.8416754603385925, | |
| "eval_matthews_correlation": 0.4605477899795795, | |
| "eval_runtime": 0.3431, | |
| "eval_samples_per_second": 3040.082, | |
| "eval_steps_per_second": 192.373, | |
| "step": 4276 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4276, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "total_flos": 65041234014708.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.4325834993588377e-05, | |
| "num_train_epochs": 2, | |
| "per_device_train_batch_size": 4, | |
| "seed": 4 | |
| } | |
| } | |