| { | |
| "best_metric": 0.029447482898831367, | |
| "best_model_checkpoint": "./teapotllm/checkpoint-606", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 606, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "em_boolean": 0.3, | |
| "em_extraction": 0.5, | |
| "em_json": 0.0, | |
| "em_qa": 0.1, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.0, | |
| "epoch": 1.0, | |
| "mean_em": 0.10975609756097561, | |
| "mean_similarity": 0.6266748062902834, | |
| "mean_word_count_diff": 3.768292682926829, | |
| "similarity_boolean": 0.7320177495479584, | |
| "similarity_extraction": 0.6650020688772201, | |
| "similarity_json": 0.6181394398212433, | |
| "similarity_qa": 0.725223034620285, | |
| "similarity_summarization": 0.7639468610286713, | |
| "similarity_unanswerable": 0.550729572184776, | |
| "word_count_diff_boolean": 1.3, | |
| "word_count_diff_extraction": 2.8, | |
| "word_count_diff_json": 1.3, | |
| "word_count_diff_qa": 5.7, | |
| "word_count_diff_summarization": 16.25, | |
| "word_count_diff_unanswerable": 3.5 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.10596980154514313, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.0797, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.03168938681483269, | |
| "eval_runtime": 5.3625, | |
| "eval_samples_per_second": 15.291, | |
| "eval_steps_per_second": 2.051, | |
| "step": 202 | |
| }, | |
| { | |
| "em_boolean": 0.5, | |
| "em_extraction": 0.5, | |
| "em_json": 0.0, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.0, | |
| "epoch": 2.0, | |
| "mean_em": 0.12195121951219512, | |
| "mean_similarity": 0.698546031944272, | |
| "mean_word_count_diff": 3.8902439024390243, | |
| "similarity_boolean": 0.805043651163578, | |
| "similarity_extraction": 0.662638445571065, | |
| "similarity_json": 0.6645591214299202, | |
| "similarity_qa": 0.8006171345710754, | |
| "similarity_summarization": 0.6312390118837357, | |
| "similarity_unanswerable": 0.6691377643299731, | |
| "word_count_diff_boolean": 1.6, | |
| "word_count_diff_extraction": 4.2, | |
| "word_count_diff_json": 1.1, | |
| "word_count_diff_qa": 5.3, | |
| "word_count_diff_summarization": 17.5, | |
| "word_count_diff_unanswerable": 3.3421052631578947 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.18695186078548431, | |
| "learning_rate": 4e-05, | |
| "loss": 0.025, | |
| "step": 404 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.259033203125, | |
| "eval_runtime": 5.3856, | |
| "eval_samples_per_second": 15.226, | |
| "eval_steps_per_second": 2.042, | |
| "step": 404 | |
| }, | |
| { | |
| "em_boolean": 0.3, | |
| "em_extraction": 0.4, | |
| "em_json": 0.0, | |
| "em_qa": 0.0, | |
| "em_summarization": 0.0, | |
| "em_unanswerable": 0.0, | |
| "epoch": 3.0, | |
| "mean_em": 0.08536585365853659, | |
| "mean_similarity": 0.6537803071664601, | |
| "mean_word_count_diff": 3.3902439024390243, | |
| "similarity_boolean": 0.6942428797483444, | |
| "similarity_extraction": 0.6155588641762734, | |
| "similarity_json": 0.7249571442604065, | |
| "similarity_qa": 0.7958889484405518, | |
| "similarity_summarization": 0.7154000997543335, | |
| "similarity_unanswerable": 0.5905764847993851, | |
| "word_count_diff_boolean": 0.9, | |
| "word_count_diff_extraction": 3.0, | |
| "word_count_diff_json": 1.4, | |
| "word_count_diff_qa": 4.3, | |
| "word_count_diff_summarization": 18.75, | |
| "word_count_diff_unanswerable": 2.8157894736842106 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.10899668186903, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.0167, | |
| "step": 606 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.029447482898831367, | |
| "eval_runtime": 5.3733, | |
| "eval_samples_per_second": 15.261, | |
| "eval_steps_per_second": 2.047, | |
| "step": 606 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2020, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 2793378756427776.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |