| { | |
| "best_metric": 2.363229513168335, | |
| "best_model_checkpoint": "bert_tiny_lda_100_v1_stsb/checkpoint-138", | |
| "epoch": 11.0, | |
| "eval_steps": 500, | |
| "global_step": 253, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.582594633102417, | |
| "learning_rate": 4.9e-05, | |
| "loss": 2.8077, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_combined_score": 0.06606349756318877, | |
| "eval_loss": 2.3849637508392334, | |
| "eval_pearson": 0.06667821062565503, | |
| "eval_runtime": 0.4882, | |
| "eval_samples_per_second": 3072.315, | |
| "eval_spearmanr": 0.06544878450072253, | |
| "eval_steps_per_second": 12.289, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 7.826333522796631, | |
| "learning_rate": 4.8e-05, | |
| "loss": 2.042, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_combined_score": 0.08412433299225583, | |
| "eval_loss": 2.608870506286621, | |
| "eval_pearson": 0.08343756630067128, | |
| "eval_runtime": 0.498, | |
| "eval_samples_per_second": 3012.31, | |
| "eval_spearmanr": 0.08481109968384036, | |
| "eval_steps_per_second": 12.049, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.146524429321289, | |
| "learning_rate": 4.7e-05, | |
| "loss": 1.9254, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_combined_score": 0.14664388177022744, | |
| "eval_loss": 2.3925955295562744, | |
| "eval_pearson": 0.15389971532281455, | |
| "eval_runtime": 0.4854, | |
| "eval_samples_per_second": 3089.992, | |
| "eval_spearmanr": 0.13938804821764034, | |
| "eval_steps_per_second": 12.36, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 8.050851821899414, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 1.8381, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_combined_score": 0.17322394096142973, | |
| "eval_loss": 2.5560152530670166, | |
| "eval_pearson": 0.17442871935710869, | |
| "eval_runtime": 0.4812, | |
| "eval_samples_per_second": 3117.34, | |
| "eval_spearmanr": 0.17201916256575076, | |
| "eval_steps_per_second": 12.469, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 35.69472122192383, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.6974, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_combined_score": 0.17843619975720487, | |
| "eval_loss": 3.0256927013397217, | |
| "eval_pearson": 0.18117857347398758, | |
| "eval_runtime": 0.4745, | |
| "eval_samples_per_second": 3161.345, | |
| "eval_spearmanr": 0.17569382604042216, | |
| "eval_steps_per_second": 12.645, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 33.44086837768555, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 1.5776, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_combined_score": 0.23191784902182225, | |
| "eval_loss": 2.363229513168335, | |
| "eval_pearson": 0.23504600421092406, | |
| "eval_runtime": 0.4766, | |
| "eval_samples_per_second": 3147.58, | |
| "eval_spearmanr": 0.22878969383272044, | |
| "eval_steps_per_second": 12.59, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 9.113436698913574, | |
| "learning_rate": 4.3e-05, | |
| "loss": 1.2951, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_combined_score": 0.25835199884600446, | |
| "eval_loss": 2.4535396099090576, | |
| "eval_pearson": 0.25938358753958196, | |
| "eval_runtime": 0.4759, | |
| "eval_samples_per_second": 3151.903, | |
| "eval_spearmanr": 0.25732041015242696, | |
| "eval_steps_per_second": 12.608, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 27.608457565307617, | |
| "learning_rate": 4.2e-05, | |
| "loss": 1.0896, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_combined_score": 0.26369469335834095, | |
| "eval_loss": 2.5246102809906006, | |
| "eval_pearson": 0.2651694385752293, | |
| "eval_runtime": 0.4775, | |
| "eval_samples_per_second": 3141.474, | |
| "eval_spearmanr": 0.26221994814145255, | |
| "eval_steps_per_second": 12.566, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 37.94071960449219, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.9372, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_combined_score": 0.26428845490221176, | |
| "eval_loss": 2.982673406600952, | |
| "eval_pearson": 0.2715918907093777, | |
| "eval_runtime": 0.482, | |
| "eval_samples_per_second": 3111.778, | |
| "eval_spearmanr": 0.25698501909504573, | |
| "eval_steps_per_second": 12.447, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 15.401611328125, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7915, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_combined_score": 0.2991297782809159, | |
| "eval_loss": 2.6918020248413086, | |
| "eval_pearson": 0.3056396011613086, | |
| "eval_runtime": 0.4829, | |
| "eval_samples_per_second": 3106.332, | |
| "eval_spearmanr": 0.29261995540052316, | |
| "eval_steps_per_second": 12.425, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 8.023918151855469, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.673, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_combined_score": 0.29500401259607845, | |
| "eval_loss": 2.752037763595581, | |
| "eval_pearson": 0.3012777534013385, | |
| "eval_runtime": 0.4793, | |
| "eval_samples_per_second": 3129.505, | |
| "eval_spearmanr": 0.2887302717908184, | |
| "eval_steps_per_second": 12.518, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "step": 253, | |
| "total_flos": 1658303025916416.0, | |
| "train_loss": 1.5158629624739937, | |
| "train_runtime": 49.0816, | |
| "train_samples_per_second": 5856.575, | |
| "train_steps_per_second": 23.43 | |
| } | |
| ], | |
| "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": 1658303025916416.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |