| { | |
| "best_metric": 0.6573427319526672, | |
| "best_model_checkpoint": "bert_base_lda_20_v1_qqp/checkpoint-9954", | |
| "epoch": 12.0, | |
| "eval_steps": 500, | |
| "global_step": 17064, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.27242642641067505, | |
| "learning_rate": 0.00098, | |
| "loss": 0.6686, | |
| "step": 1422 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.6587902903556824, | |
| "eval_runtime": 25.9114, | |
| "eval_samples_per_second": 1560.318, | |
| "eval_steps_per_second": 6.098, | |
| "step": 1422 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.06442803144454956, | |
| "learning_rate": 0.00096, | |
| "loss": 0.659, | |
| "step": 2844 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.6577603816986084, | |
| "eval_runtime": 25.983, | |
| "eval_samples_per_second": 1556.017, | |
| "eval_steps_per_second": 6.081, | |
| "step": 2844 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.05524799972772598, | |
| "learning_rate": 0.00094, | |
| "loss": 0.6586, | |
| "step": 4266 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.658168613910675, | |
| "eval_runtime": 25.9731, | |
| "eval_samples_per_second": 1556.612, | |
| "eval_steps_per_second": 6.083, | |
| "step": 4266 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.08356596529483795, | |
| "learning_rate": 0.00092, | |
| "loss": 0.6586, | |
| "step": 5688 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.6573521494865417, | |
| "eval_runtime": 25.8762, | |
| "eval_samples_per_second": 1562.439, | |
| "eval_steps_per_second": 6.106, | |
| "step": 5688 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.12229033559560776, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 0.6585, | |
| "step": 7110 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.6575562953948975, | |
| "eval_runtime": 25.8412, | |
| "eval_samples_per_second": 1564.554, | |
| "eval_steps_per_second": 6.114, | |
| "step": 7110 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.03695310652256012, | |
| "learning_rate": 0.00088, | |
| "loss": 0.6587, | |
| "step": 8532 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.6579644680023193, | |
| "eval_runtime": 25.8896, | |
| "eval_samples_per_second": 1561.63, | |
| "eval_steps_per_second": 6.103, | |
| "step": 8532 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.050063714385032654, | |
| "learning_rate": 0.00086, | |
| "loss": 0.6584, | |
| "step": 9954 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.6573427319526672, | |
| "eval_runtime": 25.8007, | |
| "eval_samples_per_second": 1567.01, | |
| "eval_steps_per_second": 6.124, | |
| "step": 9954 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.11186650395393372, | |
| "learning_rate": 0.00084, | |
| "loss": 0.6585, | |
| "step": 11376 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.6573427319526672, | |
| "eval_runtime": 25.7182, | |
| "eval_samples_per_second": 1572.04, | |
| "eval_steps_per_second": 6.144, | |
| "step": 11376 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.12498664855957031, | |
| "learning_rate": 0.00082, | |
| "loss": 0.6586, | |
| "step": 12798 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.6577603816986084, | |
| "eval_runtime": 25.6524, | |
| "eval_samples_per_second": 1576.069, | |
| "eval_steps_per_second": 6.159, | |
| "step": 12798 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.039975930005311966, | |
| "learning_rate": 0.0008, | |
| "loss": 0.6586, | |
| "step": 14220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.6577603816986084, | |
| "eval_runtime": 25.6862, | |
| "eval_samples_per_second": 1573.998, | |
| "eval_steps_per_second": 6.151, | |
| "step": 14220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.04419417306780815, | |
| "learning_rate": 0.0007800000000000001, | |
| "loss": 0.6586, | |
| "step": 15642 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.6579644680023193, | |
| "eval_runtime": 25.6827, | |
| "eval_samples_per_second": 1574.211, | |
| "eval_steps_per_second": 6.152, | |
| "step": 15642 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.040914136916399, | |
| "learning_rate": 0.00076, | |
| "loss": 0.662, | |
| "step": 17064 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6318327974276527, | |
| "eval_combined_score": 0.3159163987138264, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.6575562953948975, | |
| "eval_runtime": 25.6334, | |
| "eval_samples_per_second": 1577.242, | |
| "eval_steps_per_second": 6.164, | |
| "step": 17064 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 17064, | |
| "total_flos": 5.7439143029108736e+17, | |
| "train_loss": 0.6597107012284177, | |
| "train_runtime": 7311.2859, | |
| "train_samples_per_second": 2488.249, | |
| "train_steps_per_second": 9.725 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 71100, | |
| "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": 5.7439143029108736e+17, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |