| { | |
| "best_metric": 0.6650304198265076, | |
| "best_model_checkpoint": "bert_base_lda_20_v1_book_stsb/checkpoint-276", | |
| "epoch": 17.0, | |
| "eval_steps": 500, | |
| "global_step": 391, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 6.7477006912231445, | |
| "learning_rate": 4.9e-05, | |
| "loss": 2.8738, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_combined_score": 0.17561117616234614, | |
| "eval_loss": 2.4670398235321045, | |
| "eval_pearson": 0.17646745724560672, | |
| "eval_runtime": 0.9997, | |
| "eval_samples_per_second": 1500.502, | |
| "eval_spearmanr": 0.17475489507908556, | |
| "eval_steps_per_second": 6.002, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 35.44430160522461, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.4719, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_combined_score": 0.7400782076458396, | |
| "eval_loss": 1.0279998779296875, | |
| "eval_pearson": 0.739723981540606, | |
| "eval_runtime": 1.0078, | |
| "eval_samples_per_second": 1488.42, | |
| "eval_spearmanr": 0.7404324337510731, | |
| "eval_steps_per_second": 5.954, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 13.613875389099121, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.9801, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_combined_score": 0.7954812287252586, | |
| "eval_loss": 0.8275899887084961, | |
| "eval_pearson": 0.7955648654056936, | |
| "eval_runtime": 1.0022, | |
| "eval_samples_per_second": 1496.75, | |
| "eval_spearmanr": 0.7953975920448235, | |
| "eval_steps_per_second": 5.987, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 16.052230834960938, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.783, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_combined_score": 0.8195001768993989, | |
| "eval_loss": 0.7431384921073914, | |
| "eval_pearson": 0.8197032146737679, | |
| "eval_runtime": 0.9931, | |
| "eval_samples_per_second": 1510.427, | |
| "eval_spearmanr": 0.8192971391250299, | |
| "eval_steps_per_second": 6.042, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 12.761335372924805, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5677, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_combined_score": 0.8143535964097317, | |
| "eval_loss": 0.9074862599372864, | |
| "eval_pearson": 0.8135246219484462, | |
| "eval_runtime": 1.0017, | |
| "eval_samples_per_second": 1497.517, | |
| "eval_spearmanr": 0.8151825708710173, | |
| "eval_steps_per_second": 5.99, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 14.109099388122559, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.4407, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_combined_score": 0.8269323052226907, | |
| "eval_loss": 0.7474443316459656, | |
| "eval_pearson": 0.8267083714336028, | |
| "eval_runtime": 1.0083, | |
| "eval_samples_per_second": 1487.66, | |
| "eval_spearmanr": 0.8271562390117786, | |
| "eval_steps_per_second": 5.951, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 6.391584396362305, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.3821, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_combined_score": 0.838118234588908, | |
| "eval_loss": 0.6753207445144653, | |
| "eval_pearson": 0.8391483582280732, | |
| "eval_runtime": 0.9964, | |
| "eval_samples_per_second": 1505.478, | |
| "eval_spearmanr": 0.8370881109497429, | |
| "eval_steps_per_second": 6.022, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 9.437135696411133, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.3036, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_combined_score": 0.8252995711473009, | |
| "eval_loss": 0.8726378679275513, | |
| "eval_pearson": 0.8245684237762525, | |
| "eval_runtime": 0.9644, | |
| "eval_samples_per_second": 1555.381, | |
| "eval_spearmanr": 0.8260307185183493, | |
| "eval_steps_per_second": 6.222, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 6.426690101623535, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.269, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_combined_score": 0.8302312339843521, | |
| "eval_loss": 0.7330970168113708, | |
| "eval_pearson": 0.8311362222183013, | |
| "eval_runtime": 0.968, | |
| "eval_samples_per_second": 1549.529, | |
| "eval_spearmanr": 0.8293262457504029, | |
| "eval_steps_per_second": 6.198, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 6.680917263031006, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2191, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_combined_score": 0.8375282460597571, | |
| "eval_loss": 0.7562392354011536, | |
| "eval_pearson": 0.8382616394860685, | |
| "eval_runtime": 0.9959, | |
| "eval_samples_per_second": 1506.245, | |
| "eval_spearmanr": 0.8367948526334458, | |
| "eval_steps_per_second": 6.025, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 3.755885124206543, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.1854, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_combined_score": 0.8353770667011837, | |
| "eval_loss": 0.7022337317466736, | |
| "eval_pearson": 0.8364761178604215, | |
| "eval_runtime": 1.0014, | |
| "eval_samples_per_second": 1497.897, | |
| "eval_spearmanr": 0.834278015541946, | |
| "eval_steps_per_second": 5.992, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 7.932562828063965, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.1718, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_combined_score": 0.8394422302882345, | |
| "eval_loss": 0.6650304198265076, | |
| "eval_pearson": 0.8406845213042464, | |
| "eval_runtime": 0.9788, | |
| "eval_samples_per_second": 1532.489, | |
| "eval_spearmanr": 0.8381999392722225, | |
| "eval_steps_per_second": 6.13, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 6.083053112030029, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.1685, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_combined_score": 0.8341740484470688, | |
| "eval_loss": 0.7269710302352905, | |
| "eval_pearson": 0.8350427176753902, | |
| "eval_runtime": 0.9854, | |
| "eval_samples_per_second": 1522.222, | |
| "eval_spearmanr": 0.8333053792187473, | |
| "eval_steps_per_second": 6.089, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 5.380353927612305, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.1368, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_combined_score": 0.8384150656102521, | |
| "eval_loss": 0.7532300353050232, | |
| "eval_pearson": 0.839212923010702, | |
| "eval_runtime": 0.9703, | |
| "eval_samples_per_second": 1545.963, | |
| "eval_spearmanr": 0.8376172082098022, | |
| "eval_steps_per_second": 6.184, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 2.213524103164673, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.1351, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_combined_score": 0.8378988273941752, | |
| "eval_loss": 0.8710149526596069, | |
| "eval_pearson": 0.8378603538577427, | |
| "eval_runtime": 0.9708, | |
| "eval_samples_per_second": 1545.192, | |
| "eval_spearmanr": 0.8379373009306077, | |
| "eval_steps_per_second": 6.181, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 5.186318874359131, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.1459, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_combined_score": 0.840746533861038, | |
| "eval_loss": 0.7801129817962646, | |
| "eval_pearson": 0.841643298465437, | |
| "eval_runtime": 0.9846, | |
| "eval_samples_per_second": 1523.477, | |
| "eval_spearmanr": 0.8398497692566391, | |
| "eval_steps_per_second": 6.094, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 4.310553550720215, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.106, | |
| "step": 391 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_combined_score": 0.8386885053587652, | |
| "eval_loss": 0.6833200454711914, | |
| "eval_pearson": 0.8393476630897322, | |
| "eval_runtime": 1.0147, | |
| "eval_samples_per_second": 1478.234, | |
| "eval_spearmanr": 0.838029347627798, | |
| "eval_steps_per_second": 5.913, | |
| "step": 391 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "step": 391, | |
| "total_flos": 1.2857200946093568e+16, | |
| "train_loss": 0.5494434967675172, | |
| "train_runtime": 221.1064, | |
| "train_samples_per_second": 1300.053, | |
| "train_steps_per_second": 5.201 | |
| } | |
| ], | |
| "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": 1.2857200946093568e+16, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |