| { | |
| "best_metric": 0.5353626178321323, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-6/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": 0.9345794392523364, | |
| "grad_norm": 4.85833215713501, | |
| "learning_rate": 3.693802167824332e-05, | |
| "loss": 0.5213, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.48894789814949036, | |
| "eval_matthews_correlation": 0.41310412373921845, | |
| "eval_runtime": 0.7397, | |
| "eval_samples_per_second": 1410.045, | |
| "eval_steps_per_second": 89.226, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 1.8691588785046729, | |
| "grad_norm": 12.77253532409668, | |
| "learning_rate": 2.844652244186554e-05, | |
| "loss": 0.3193, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5021252036094666, | |
| "eval_matthews_correlation": 0.5324104093270398, | |
| "eval_runtime": 0.835, | |
| "eval_samples_per_second": 1249.081, | |
| "eval_steps_per_second": 79.041, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.803738317757009, | |
| "grad_norm": 0.39278796315193176, | |
| "learning_rate": 1.9955023205487768e-05, | |
| "loss": 0.219, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.808618426322937, | |
| "eval_matthews_correlation": 0.5035428557399136, | |
| "eval_runtime": 0.8341, | |
| "eval_samples_per_second": 1250.464, | |
| "eval_steps_per_second": 79.128, | |
| "step": 1605 | |
| }, | |
| { | |
| "epoch": 3.7383177570093458, | |
| "grad_norm": 38.36227035522461, | |
| "learning_rate": 1.1463523969109994e-05, | |
| "loss": 0.1254, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8212687969207764, | |
| "eval_matthews_correlation": 0.5353626178321323, | |
| "eval_runtime": 0.808, | |
| "eval_samples_per_second": 1290.873, | |
| "eval_steps_per_second": 81.685, | |
| "step": 2140 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2675, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 176447609911344.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 4.542952091462109e-05, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 16, | |
| "seed": 10 | |
| } | |
| } | |