| { |
| "best_global_step": 4266, |
| "best_metric": 0.4080028235912323, |
| "best_model_checkpoint": "tiny_bert_rand_5_v1_qqp/checkpoint-4266", |
| "epoch": 8.0, |
| "eval_steps": 500, |
| "global_step": 11376, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 2.0584754943847656, |
| "learning_rate": 4.9e-05, |
| "loss": 0.4808, |
| "step": 1422 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.789463269849122, |
| "eval_combined_score": 0.7368681334408815, |
| "eval_f1": 0.684272997032641, |
| "eval_loss": 0.4439510107040405, |
| "eval_runtime": 11.9948, |
| "eval_samples_per_second": 3370.641, |
| "eval_steps_per_second": 13.172, |
| "step": 1422 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 2.4025912284851074, |
| "learning_rate": 4.8e-05, |
| "loss": 0.3876, |
| "step": 2844 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8119465743259956, |
| "eval_combined_score": 0.7726855342791579, |
| "eval_f1": 0.73342449423232, |
| "eval_loss": 0.4098320007324219, |
| "eval_runtime": 11.777, |
| "eval_samples_per_second": 3432.955, |
| "eval_steps_per_second": 13.416, |
| "step": 2844 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 2.9840660095214844, |
| "learning_rate": 4.7e-05, |
| "loss": 0.328, |
| "step": 4266 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8182785060598565, |
| "eval_combined_score": 0.7856145884480661, |
| "eval_f1": 0.7529506708362756, |
| "eval_loss": 0.4080028235912323, |
| "eval_runtime": 11.7624, |
| "eval_samples_per_second": 3437.227, |
| "eval_steps_per_second": 13.433, |
| "step": 4266 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 3.496941089630127, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.28, |
| "step": 5688 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8232005936186001, |
| "eval_combined_score": 0.7838479376443015, |
| "eval_f1": 0.7444952816700029, |
| "eval_loss": 0.4340234100818634, |
| "eval_runtime": 11.8403, |
| "eval_samples_per_second": 3414.608, |
| "eval_steps_per_second": 13.344, |
| "step": 5688 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 3.7349326610565186, |
| "learning_rate": 4.5e-05, |
| "loss": 0.2399, |
| "step": 7110 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.830348750927529, |
| "eval_combined_score": 0.7952324316313579, |
| "eval_f1": 0.7601161123351869, |
| "eval_loss": 0.4672740697860718, |
| "eval_runtime": 11.7579, |
| "eval_samples_per_second": 3438.542, |
| "eval_steps_per_second": 13.438, |
| "step": 7110 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 3.271902322769165, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.2063, |
| "step": 8532 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8257976749938165, |
| "eval_combined_score": 0.7954646160009267, |
| "eval_f1": 0.7651315570080368, |
| "eval_loss": 0.46120214462280273, |
| "eval_runtime": 11.8391, |
| "eval_samples_per_second": 3414.95, |
| "eval_steps_per_second": 13.346, |
| "step": 8532 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 4.440460205078125, |
| "learning_rate": 4.3e-05, |
| "loss": 0.1799, |
| "step": 9954 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8302992827108583, |
| "eval_combined_score": 0.8005896146397837, |
| "eval_f1": 0.7708799465687093, |
| "eval_loss": 0.5016512870788574, |
| "eval_runtime": 11.8725, |
| "eval_samples_per_second": 3405.362, |
| "eval_steps_per_second": 13.308, |
| "step": 9954 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 4.346182346343994, |
| "learning_rate": 4.2e-05, |
| "loss": 0.1581, |
| "step": 11376 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8344298788028691, |
| "eval_combined_score": 0.8029516593249632, |
| "eval_f1": 0.7714734398470572, |
| "eval_loss": 0.5428574681282043, |
| "eval_runtime": 11.8689, |
| "eval_samples_per_second": 3406.377, |
| "eval_steps_per_second": 13.312, |
| "step": 11376 |
| }, |
| { |
| "epoch": 8.0, |
| "step": 11376, |
| "total_flos": 7.633075201391002e+16, |
| "train_loss": 0.2825857328798533, |
| "train_runtime": 1627.1899, |
| "train_samples_per_second": 11180.195, |
| "train_steps_per_second": 43.695 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 71100, |
| "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": 7.633075201391002e+16, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
