| { | |
| "best_metric": 0.8188873529434204, | |
| "best_model_checkpoint": "bert_tiny_lda_5_v1_book_stsb/checkpoint-368", | |
| "epoch": 21.0, | |
| "eval_steps": 500, | |
| "global_step": 483, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.9694929122924805, | |
| "learning_rate": 4.9e-05, | |
| "loss": 2.8706, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_combined_score": 0.09954146094056157, | |
| "eval_loss": 2.462472438812256, | |
| "eval_pearson": 0.09907601559786483, | |
| "eval_runtime": 0.4799, | |
| "eval_samples_per_second": 3125.334, | |
| "eval_spearmanr": 0.1000069062832583, | |
| "eval_steps_per_second": 12.501, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 22.507034301757812, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.7766, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_combined_score": 0.6395870539114769, | |
| "eval_loss": 1.3386552333831787, | |
| "eval_pearson": 0.6420563993485077, | |
| "eval_runtime": 0.5098, | |
| "eval_samples_per_second": 2942.208, | |
| "eval_spearmanr": 0.637117708474446, | |
| "eval_steps_per_second": 11.769, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 20.13024139404297, | |
| "learning_rate": 4.7e-05, | |
| "loss": 1.1649, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_combined_score": 0.7499948802864915, | |
| "eval_loss": 0.9883082509040833, | |
| "eval_pearson": 0.7530220350451059, | |
| "eval_runtime": 0.4998, | |
| "eval_samples_per_second": 3001.186, | |
| "eval_spearmanr": 0.746967725527877, | |
| "eval_steps_per_second": 12.005, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 7.193154335021973, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.8446, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_combined_score": 0.7814449278644269, | |
| "eval_loss": 0.9500728249549866, | |
| "eval_pearson": 0.7801592206629702, | |
| "eval_runtime": 0.4748, | |
| "eval_samples_per_second": 3159.192, | |
| "eval_spearmanr": 0.7827306350658837, | |
| "eval_steps_per_second": 12.637, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 15.801517486572266, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6587, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_combined_score": 0.7831183677075164, | |
| "eval_loss": 0.9120289087295532, | |
| "eval_pearson": 0.7838836623851876, | |
| "eval_runtime": 0.4934, | |
| "eval_samples_per_second": 3040.098, | |
| "eval_spearmanr": 0.7823530730298451, | |
| "eval_steps_per_second": 12.16, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 14.146763801574707, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.5587, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_combined_score": 0.7878115431321093, | |
| "eval_loss": 0.8917849659919739, | |
| "eval_pearson": 0.7876118713697591, | |
| "eval_runtime": 0.4812, | |
| "eval_samples_per_second": 3117.386, | |
| "eval_spearmanr": 0.7880112148944596, | |
| "eval_steps_per_second": 12.47, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 12.302091598510742, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.5104, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_combined_score": 0.782515489635635, | |
| "eval_loss": 0.9523068070411682, | |
| "eval_pearson": 0.7821895116539266, | |
| "eval_runtime": 0.4809, | |
| "eval_samples_per_second": 3119.023, | |
| "eval_spearmanr": 0.7828414676173435, | |
| "eval_steps_per_second": 12.476, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 12.979435920715332, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.4506, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_combined_score": 0.7919260991828585, | |
| "eval_loss": 0.9420682191848755, | |
| "eval_pearson": 0.791584489351049, | |
| "eval_runtime": 0.4809, | |
| "eval_samples_per_second": 3118.926, | |
| "eval_spearmanr": 0.7922677090146679, | |
| "eval_steps_per_second": 12.476, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 18.948219299316406, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.4174, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_combined_score": 0.7906456677090563, | |
| "eval_loss": 1.0924893617630005, | |
| "eval_pearson": 0.7875795388915195, | |
| "eval_runtime": 0.4739, | |
| "eval_samples_per_second": 3165.183, | |
| "eval_spearmanr": 0.7937117965265932, | |
| "eval_steps_per_second": 12.661, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 5.4617791175842285, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3229, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_combined_score": 0.7957522599726154, | |
| "eval_loss": 0.8792051076889038, | |
| "eval_pearson": 0.7969462774842241, | |
| "eval_runtime": 0.4767, | |
| "eval_samples_per_second": 3146.57, | |
| "eval_spearmanr": 0.7945582424610067, | |
| "eval_steps_per_second": 12.586, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 6.967871189117432, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.3096, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_combined_score": 0.7885789869359312, | |
| "eval_loss": 0.9665384292602539, | |
| "eval_pearson": 0.7887171688488032, | |
| "eval_runtime": 0.487, | |
| "eval_samples_per_second": 3079.987, | |
| "eval_spearmanr": 0.7884408050230591, | |
| "eval_steps_per_second": 12.32, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 16.113603591918945, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.2918, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_combined_score": 0.8004322602860193, | |
| "eval_loss": 0.8212088346481323, | |
| "eval_pearson": 0.801677826027752, | |
| "eval_runtime": 0.4993, | |
| "eval_samples_per_second": 3003.968, | |
| "eval_spearmanr": 0.7991866945442867, | |
| "eval_steps_per_second": 12.016, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 4.451534271240234, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.2479, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_combined_score": 0.804614899302114, | |
| "eval_loss": 0.8791233897209167, | |
| "eval_pearson": 0.8055416707360773, | |
| "eval_runtime": 0.4721, | |
| "eval_samples_per_second": 3177.294, | |
| "eval_spearmanr": 0.8036881278681508, | |
| "eval_steps_per_second": 12.709, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 21.7775821685791, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.2409, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_combined_score": 0.794995597443876, | |
| "eval_loss": 0.8217026591300964, | |
| "eval_pearson": 0.7969938156234373, | |
| "eval_runtime": 0.5148, | |
| "eval_samples_per_second": 2913.928, | |
| "eval_spearmanr": 0.7929973792643148, | |
| "eval_steps_per_second": 11.656, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 3.07222056388855, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2071, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_combined_score": 0.7984497715007259, | |
| "eval_loss": 0.8929257988929749, | |
| "eval_pearson": 0.7998156602749035, | |
| "eval_runtime": 0.4892, | |
| "eval_samples_per_second": 3066.293, | |
| "eval_spearmanr": 0.7970838827265485, | |
| "eval_steps_per_second": 12.265, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 3.4134459495544434, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.1907, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_combined_score": 0.8007289150909647, | |
| "eval_loss": 0.8188873529434204, | |
| "eval_pearson": 0.8026761278764143, | |
| "eval_runtime": 0.4814, | |
| "eval_samples_per_second": 3116.2, | |
| "eval_spearmanr": 0.798781702305515, | |
| "eval_steps_per_second": 12.465, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 4.4020676612854, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.1825, | |
| "step": 391 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_combined_score": 0.79118899198846, | |
| "eval_loss": 0.9392598271369934, | |
| "eval_pearson": 0.7924365508807105, | |
| "eval_runtime": 0.4766, | |
| "eval_samples_per_second": 3147.292, | |
| "eval_spearmanr": 0.7899414330962096, | |
| "eval_steps_per_second": 12.589, | |
| "step": 391 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 5.791975498199463, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.1634, | |
| "step": 414 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_combined_score": 0.8034539561261447, | |
| "eval_loss": 0.8721528053283691, | |
| "eval_pearson": 0.8045915853539458, | |
| "eval_runtime": 0.4845, | |
| "eval_samples_per_second": 3095.701, | |
| "eval_spearmanr": 0.8023163268983434, | |
| "eval_steps_per_second": 12.383, | |
| "step": 414 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 2.9870147705078125, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.1531, | |
| "step": 437 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_combined_score": 0.8030149651947462, | |
| "eval_loss": 0.9629133343696594, | |
| "eval_pearson": 0.8037084608596774, | |
| "eval_runtime": 0.4776, | |
| "eval_samples_per_second": 3140.565, | |
| "eval_spearmanr": 0.8023214695298152, | |
| "eval_steps_per_second": 12.562, | |
| "step": 437 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.91269850730896, | |
| "learning_rate": 3e-05, | |
| "loss": 0.153, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_combined_score": 0.7958772509210114, | |
| "eval_loss": 1.0062919855117798, | |
| "eval_pearson": 0.797094458559933, | |
| "eval_runtime": 0.485, | |
| "eval_samples_per_second": 3092.717, | |
| "eval_spearmanr": 0.7946600432820897, | |
| "eval_steps_per_second": 12.371, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "grad_norm": 4.98586893081665, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.1409, | |
| "step": 483 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_combined_score": 0.7955185663928932, | |
| "eval_loss": 0.9756875038146973, | |
| "eval_pearson": 0.7968145893338887, | |
| "eval_runtime": 0.478, | |
| "eval_samples_per_second": 3138.129, | |
| "eval_spearmanr": 0.7942225434518977, | |
| "eval_steps_per_second": 12.553, | |
| "step": 483 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "step": 483, | |
| "total_flos": 3165851231294976.0, | |
| "train_loss": 0.5645940970189823, | |
| "train_runtime": 97.1506, | |
| "train_samples_per_second": 2958.809, | |
| "train_steps_per_second": 11.837 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 1150, | |
| "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": 3165851231294976.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |