| { | |
| "best_global_step": 69, | |
| "best_metric": 2.357128858566284, | |
| "best_model_checkpoint": "tiny_bert_rand_100_v2_stsb/checkpoint-69", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 184, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 7.353635311126709, | |
| "learning_rate": 4.9e-05, | |
| "loss": 3.0934, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_combined_score": 0.1262982832568416, | |
| "eval_loss": 2.4178946018218994, | |
| "eval_pearson": 0.1274026693406406, | |
| "eval_runtime": 0.4494, | |
| "eval_samples_per_second": 3337.766, | |
| "eval_spearmanr": 0.12519389717304258, | |
| "eval_steps_per_second": 13.351, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.9686944484710693, | |
| "learning_rate": 4.8e-05, | |
| "loss": 2.0262, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_combined_score": 0.08032337810098913, | |
| "eval_loss": 2.822695255279541, | |
| "eval_pearson": 0.09062722453006844, | |
| "eval_runtime": 0.483, | |
| "eval_samples_per_second": 3105.343, | |
| "eval_spearmanr": 0.07001953167190983, | |
| "eval_steps_per_second": 12.421, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.415579319000244, | |
| "learning_rate": 4.7e-05, | |
| "loss": 1.8632, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_combined_score": 0.18249862203222653, | |
| "eval_loss": 2.357128858566284, | |
| "eval_pearson": 0.19035307550832772, | |
| "eval_runtime": 0.4733, | |
| "eval_samples_per_second": 3168.942, | |
| "eval_spearmanr": 0.17464416855612533, | |
| "eval_steps_per_second": 12.676, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 6.482253551483154, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 1.6504, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_combined_score": 0.23817120921581703, | |
| "eval_loss": 2.4674251079559326, | |
| "eval_pearson": 0.24049226357258544, | |
| "eval_runtime": 0.4467, | |
| "eval_samples_per_second": 3358.126, | |
| "eval_spearmanr": 0.23585015485904862, | |
| "eval_steps_per_second": 13.433, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 7.711766719818115, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.376, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_combined_score": 0.24243941396924645, | |
| "eval_loss": 2.41092848777771, | |
| "eval_pearson": 0.24433716723818158, | |
| "eval_runtime": 0.4757, | |
| "eval_samples_per_second": 3153.427, | |
| "eval_spearmanr": 0.2405416607003113, | |
| "eval_steps_per_second": 12.614, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 7.9283013343811035, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 1.1686, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_combined_score": 0.2585957948418378, | |
| "eval_loss": 2.5537660121917725, | |
| "eval_pearson": 0.2572703578545987, | |
| "eval_runtime": 0.46, | |
| "eval_samples_per_second": 3260.655, | |
| "eval_spearmanr": 0.2599212318290769, | |
| "eval_steps_per_second": 13.043, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 8.92582893371582, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.9782, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_combined_score": 0.26390896222736915, | |
| "eval_loss": 2.6227357387542725, | |
| "eval_pearson": 0.2622185613208425, | |
| "eval_runtime": 0.489, | |
| "eval_samples_per_second": 3067.175, | |
| "eval_spearmanr": 0.2655993631338958, | |
| "eval_steps_per_second": 12.269, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 16.892637252807617, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.8135, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_combined_score": 0.234079183599613, | |
| "eval_loss": 3.0192532539367676, | |
| "eval_pearson": 0.2304551048967358, | |
| "eval_runtime": 0.4608, | |
| "eval_samples_per_second": 3255.012, | |
| "eval_spearmanr": 0.23770326230249023, | |
| "eval_steps_per_second": 13.02, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 184, | |
| "total_flos": 1206038564302848.0, | |
| "train_loss": 1.6211834264838176, | |
| "train_runtime": 33.9359, | |
| "train_samples_per_second": 8470.39, | |
| "train_steps_per_second": 33.887 | |
| } | |
| ], | |
| "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": 1206038564302848.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |