| { | |
| "best_metric": 0.5245014079698576, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-6/checkpoint-6414", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 6414, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 17.90995979309082, | |
| "learning_rate": 1.8573493164868257e-05, | |
| "loss": 0.6076, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 25.591594696044922, | |
| "learning_rate": 1.700319445292471e-05, | |
| "loss": 0.5721, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 17.508258819580078, | |
| "learning_rate": 1.5432895740981165e-05, | |
| "loss": 0.5514, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 47.17720031738281, | |
| "learning_rate": 1.3862597029037618e-05, | |
| "loss": 0.5529, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.4856245219707489, | |
| "eval_matthews_correlation": 0.47921629302778096, | |
| "eval_runtime": 0.8291, | |
| "eval_samples_per_second": 1258.047, | |
| "eval_steps_per_second": 79.608, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 0.4112852215766907, | |
| "learning_rate": 1.2292298317094074e-05, | |
| "loss": 0.4362, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 34.63631057739258, | |
| "learning_rate": 1.0721999605150526e-05, | |
| "loss": 0.435, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 0.18024270236492157, | |
| "learning_rate": 9.151700893206982e-06, | |
| "loss": 0.4341, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 0.2709442675113678, | |
| "learning_rate": 7.581402181263437e-06, | |
| "loss": 0.4354, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.7808897495269775, | |
| "eval_matthews_correlation": 0.4758828120362657, | |
| "eval_runtime": 0.7521, | |
| "eval_samples_per_second": 1386.711, | |
| "eval_steps_per_second": 87.75, | |
| "step": 4276 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 68.71192169189453, | |
| "learning_rate": 6.011103469319891e-06, | |
| "loss": 0.3867, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 0.26191192865371704, | |
| "learning_rate": 4.440804757376346e-06, | |
| "loss": 0.2848, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 0.26115551590919495, | |
| "learning_rate": 2.8705060454328013e-06, | |
| "loss": 0.2901, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 0.08103559166193008, | |
| "learning_rate": 1.3002073334892556e-06, | |
| "loss": 0.281, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.9468569755554199, | |
| "eval_matthews_correlation": 0.5245014079698576, | |
| "eval_runtime": 0.7366, | |
| "eval_samples_per_second": 1415.874, | |
| "eval_steps_per_second": 89.595, | |
| "step": 6414 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6414, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 149328014741184.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 2.01437918768118e-05, | |
| "num_train_epochs": 3, | |
| "per_device_train_batch_size": 4, | |
| "seed": 9 | |
| } | |
| } | |