| { | |
| "best_metric": 0.2878433167934418, | |
| "best_model_checkpoint": "distilbert_lda_5_v1_qqp/checkpoint-4266", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 11376, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.2275643348693848, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.376, | |
| "step": 1422 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8607964382883997, | |
| "eval_combined_score": 0.8375881927589756, | |
| "eval_f1": 0.8143799472295514, | |
| "eval_loss": 0.30836477875709534, | |
| "eval_runtime": 17.0485, | |
| "eval_samples_per_second": 2371.471, | |
| "eval_steps_per_second": 9.268, | |
| "step": 1422 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.772549629211426, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.267, | |
| "step": 2844 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8694286420974524, | |
| "eval_combined_score": 0.8528312746705657, | |
| "eval_f1": 0.8362339072436792, | |
| "eval_loss": 0.2947133481502533, | |
| "eval_runtime": 17.0395, | |
| "eval_samples_per_second": 2372.726, | |
| "eval_steps_per_second": 9.273, | |
| "step": 2844 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.4764814376831055, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.2049, | |
| "step": 4266 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8815978233984665, | |
| "eval_combined_score": 0.8632012459511379, | |
| "eval_f1": 0.8448046685038093, | |
| "eval_loss": 0.2878433167934418, | |
| "eval_runtime": 17.0226, | |
| "eval_samples_per_second": 2375.077, | |
| "eval_steps_per_second": 9.282, | |
| "step": 4266 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.1702332496643066, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.1541, | |
| "step": 5688 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8844669799653723, | |
| "eval_combined_score": 0.8617960868485377, | |
| "eval_f1": 0.839125193731703, | |
| "eval_loss": 0.3069991171360016, | |
| "eval_runtime": 16.9537, | |
| "eval_samples_per_second": 2384.735, | |
| "eval_steps_per_second": 9.32, | |
| "step": 5688 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.335996150970459, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.1166, | |
| "step": 7110 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8864209745238684, | |
| "eval_combined_score": 0.8652584264213012, | |
| "eval_f1": 0.8440958783187342, | |
| "eval_loss": 0.3457874059677124, | |
| "eval_runtime": 16.9631, | |
| "eval_samples_per_second": 2383.415, | |
| "eval_steps_per_second": 9.314, | |
| "step": 7110 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.3953769207000732, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.0894, | |
| "step": 8532 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8852584714321049, | |
| "eval_combined_score": 0.865072142059049, | |
| "eval_f1": 0.8448858126859933, | |
| "eval_loss": 0.384445458650589, | |
| "eval_runtime": 17.0221, | |
| "eval_samples_per_second": 2375.142, | |
| "eval_steps_per_second": 9.282, | |
| "step": 8532 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 4.227119445800781, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.0719, | |
| "step": 9954 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8848379915904032, | |
| "eval_combined_score": 0.8664943985738853, | |
| "eval_f1": 0.8481508055573674, | |
| "eval_loss": 0.4213469326496124, | |
| "eval_runtime": 17.1944, | |
| "eval_samples_per_second": 2351.341, | |
| "eval_steps_per_second": 9.189, | |
| "step": 9954 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.2477705478668213, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.0579, | |
| "step": 11376 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8842443729903537, | |
| "eval_combined_score": 0.8655013613478291, | |
| "eval_f1": 0.8467583497053045, | |
| "eval_loss": 0.42570531368255615, | |
| "eval_runtime": 17.1364, | |
| "eval_samples_per_second": 2359.308, | |
| "eval_steps_per_second": 9.22, | |
| "step": 11376 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 11376, | |
| "total_flos": 1.927909325255639e+17, | |
| "train_loss": 0.1672345508838337, | |
| "train_runtime": 2845.5428, | |
| "train_samples_per_second": 6393.262, | |
| "train_steps_per_second": 24.986 | |
| } | |
| ], | |
| "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": 1.927909325255639e+17, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |