| { |
| "best_global_step": 68, |
| "best_metric": 0.6170597076416016, |
| "best_model_checkpoint": "distilbert_km_50_v1_cola/checkpoint-68", |
| "epoch": 7.0, |
| "eval_steps": 500, |
| "global_step": 238, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.309290885925293, |
| "learning_rate": 4.9e-05, |
| "loss": 0.6173, |
| "step": 34 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6912751793861389, |
| "eval_loss": 0.6217023730278015, |
| "eval_matthews_correlation": 0.0, |
| "eval_runtime": 0.457, |
| "eval_samples_per_second": 2282.352, |
| "eval_steps_per_second": 10.941, |
| "step": 34 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.558940052986145, |
| "learning_rate": 4.8e-05, |
| "loss": 0.6, |
| "step": 68 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.6922339200973511, |
| "eval_loss": 0.6170597076416016, |
| "eval_matthews_correlation": 0.04161454566059593, |
| "eval_runtime": 0.4485, |
| "eval_samples_per_second": 2325.316, |
| "eval_steps_per_second": 11.147, |
| "step": 68 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.7420350313186646, |
| "learning_rate": 4.7e-05, |
| "loss": 0.567, |
| "step": 102 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.693192720413208, |
| "eval_loss": 0.6260364651679993, |
| "eval_matthews_correlation": 0.11120632204905952, |
| "eval_runtime": 0.4542, |
| "eval_samples_per_second": 2296.485, |
| "eval_steps_per_second": 11.009, |
| "step": 102 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 2.4721972942352295, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.5196, |
| "step": 136 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.6864812970161438, |
| "eval_loss": 0.6525582671165466, |
| "eval_matthews_correlation": 0.10436722000562923, |
| "eval_runtime": 0.4667, |
| "eval_samples_per_second": 2234.8, |
| "eval_steps_per_second": 10.713, |
| "step": 136 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 2.3430979251861572, |
| "learning_rate": 4.5e-05, |
| "loss": 0.4429, |
| "step": 170 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6605944633483887, |
| "eval_loss": 0.6836659908294678, |
| "eval_matthews_correlation": 0.15036625003655318, |
| "eval_runtime": 0.4566, |
| "eval_samples_per_second": 2284.138, |
| "eval_steps_per_second": 10.95, |
| "step": 170 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 3.205092430114746, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.3773, |
| "step": 204 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.6558005809783936, |
| "eval_loss": 0.7564343810081482, |
| "eval_matthews_correlation": 0.12227142135867913, |
| "eval_runtime": 0.4817, |
| "eval_samples_per_second": 2165.112, |
| "eval_steps_per_second": 10.379, |
| "step": 204 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 3.901951551437378, |
| "learning_rate": 4.3e-05, |
| "loss": 0.31, |
| "step": 238 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.6452540755271912, |
| "eval_loss": 0.8317585587501526, |
| "eval_matthews_correlation": 0.12496475440802036, |
| "eval_runtime": 0.4401, |
| "eval_samples_per_second": 2370.083, |
| "eval_steps_per_second": 11.362, |
| "step": 238 |
| }, |
| { |
| "epoch": 7.0, |
| "step": 238, |
| "total_flos": 3964550540676096.0, |
| "train_loss": 0.49057099799148174, |
| "train_runtime": 68.034, |
| "train_samples_per_second": 6284.357, |
| "train_steps_per_second": 24.988 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 1700, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 50, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 5, |
| "early_stopping_threshold": 0.0 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 5 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3964550540676096.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
