| { | |
| "best_metric": 0.7980738282203674, | |
| "best_model_checkpoint": "distilbert_lda_100_v1_book_stsb/checkpoint-230", | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 345, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 13.065649032592773, | |
| "learning_rate": 4.9e-05, | |
| "loss": 3.1376, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_combined_score": 0.1728512599113079, | |
| "eval_loss": 2.3443872928619385, | |
| "eval_pearson": 0.18002362677750725, | |
| "eval_runtime": 0.4619, | |
| "eval_samples_per_second": 3247.478, | |
| "eval_spearmanr": 0.16567889304510858, | |
| "eval_steps_per_second": 12.99, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 30.20730209350586, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.571, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_combined_score": 0.6510883490332047, | |
| "eval_loss": 1.4976587295532227, | |
| "eval_pearson": 0.6469347808355328, | |
| "eval_runtime": 0.468, | |
| "eval_samples_per_second": 3205.005, | |
| "eval_spearmanr": 0.6552419172308765, | |
| "eval_steps_per_second": 12.82, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 9.512760162353516, | |
| "learning_rate": 4.7e-05, | |
| "loss": 1.0298, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_combined_score": 0.7472320299511837, | |
| "eval_loss": 0.9940208196640015, | |
| "eval_pearson": 0.7482802487477433, | |
| "eval_runtime": 0.5042, | |
| "eval_samples_per_second": 2975.284, | |
| "eval_spearmanr": 0.7461838111546238, | |
| "eval_steps_per_second": 11.901, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 9.95678424835205, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.8795, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_combined_score": 0.7665966692630239, | |
| "eval_loss": 1.0649274587631226, | |
| "eval_pearson": 0.7621958495107538, | |
| "eval_runtime": 0.4869, | |
| "eval_samples_per_second": 3080.609, | |
| "eval_spearmanr": 0.770997489015294, | |
| "eval_steps_per_second": 12.322, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 29.694782257080078, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6951, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_combined_score": 0.7678001045986155, | |
| "eval_loss": 1.5035641193389893, | |
| "eval_pearson": 0.750753453158977, | |
| "eval_runtime": 0.4839, | |
| "eval_samples_per_second": 3099.586, | |
| "eval_spearmanr": 0.7848467560382539, | |
| "eval_steps_per_second": 12.398, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 15.514448165893555, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.5558, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_combined_score": 0.7895822730889075, | |
| "eval_loss": 0.9067298769950867, | |
| "eval_pearson": 0.7877625518399618, | |
| "eval_runtime": 0.4716, | |
| "eval_samples_per_second": 3180.754, | |
| "eval_spearmanr": 0.7914019943378532, | |
| "eval_steps_per_second": 12.723, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 7.108041286468506, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.4306, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_combined_score": 0.8044942171535114, | |
| "eval_loss": 0.8333074450492859, | |
| "eval_pearson": 0.8050894731664774, | |
| "eval_runtime": 0.4837, | |
| "eval_samples_per_second": 3100.907, | |
| "eval_spearmanr": 0.8038989611405454, | |
| "eval_steps_per_second": 12.404, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.6502273082733154, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.3592, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_combined_score": 0.797110517540158, | |
| "eval_loss": 0.9581788778305054, | |
| "eval_pearson": 0.7966780577453428, | |
| "eval_runtime": 0.4867, | |
| "eval_samples_per_second": 3081.961, | |
| "eval_spearmanr": 0.7975429773349733, | |
| "eval_steps_per_second": 12.328, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 9.489797592163086, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.2847, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_combined_score": 0.7941543139172129, | |
| "eval_loss": 1.0402296781539917, | |
| "eval_pearson": 0.7929483799420015, | |
| "eval_runtime": 0.4618, | |
| "eval_samples_per_second": 3248.43, | |
| "eval_spearmanr": 0.7953602478924242, | |
| "eval_steps_per_second": 12.994, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 5.137160778045654, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2689, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_combined_score": 0.8037144170123958, | |
| "eval_loss": 0.7980738282203674, | |
| "eval_pearson": 0.8060121051059546, | |
| "eval_runtime": 0.4835, | |
| "eval_samples_per_second": 3102.36, | |
| "eval_spearmanr": 0.8014167289188371, | |
| "eval_steps_per_second": 12.409, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 9.64781665802002, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.2368, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_combined_score": 0.8091608382603208, | |
| "eval_loss": 0.8627774119377136, | |
| "eval_pearson": 0.8100673949446242, | |
| "eval_runtime": 0.4681, | |
| "eval_samples_per_second": 3204.761, | |
| "eval_spearmanr": 0.8082542815760174, | |
| "eval_steps_per_second": 12.819, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 7.976446151733398, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.2088, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_combined_score": 0.8001157143831301, | |
| "eval_loss": 1.0528961420059204, | |
| "eval_pearson": 0.7990927893008337, | |
| "eval_runtime": 0.4926, | |
| "eval_samples_per_second": 3044.855, | |
| "eval_spearmanr": 0.8011386394654265, | |
| "eval_steps_per_second": 12.179, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 9.004283905029297, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.1912, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_combined_score": 0.8012021697636892, | |
| "eval_loss": 0.8877846598625183, | |
| "eval_pearson": 0.8010653292046944, | |
| "eval_runtime": 0.4576, | |
| "eval_samples_per_second": 3278.316, | |
| "eval_spearmanr": 0.801339010322684, | |
| "eval_steps_per_second": 13.113, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 8.706100463867188, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.1618, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_combined_score": 0.7950920104148684, | |
| "eval_loss": 0.8756589293479919, | |
| "eval_pearson": 0.7958563730475406, | |
| "eval_runtime": 0.4905, | |
| "eval_samples_per_second": 3058.076, | |
| "eval_spearmanr": 0.7943276477821963, | |
| "eval_steps_per_second": 12.232, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 9.078815460205078, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.1557, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_combined_score": 0.7990363023147462, | |
| "eval_loss": 0.8970615267753601, | |
| "eval_pearson": 0.8001497386998061, | |
| "eval_runtime": 0.4666, | |
| "eval_samples_per_second": 3215.038, | |
| "eval_spearmanr": 0.7979228659296864, | |
| "eval_steps_per_second": 12.86, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 345, | |
| "total_flos": 5711561202147840.0, | |
| "train_loss": 0.6777657121851824, | |
| "train_runtime": 82.4709, | |
| "train_samples_per_second": 3485.471, | |
| "train_steps_per_second": 13.944 | |
| } | |
| ], | |
| "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": 5711561202147840.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |