| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 9956, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2008838891120932, | |
| "grad_norm": 2.296739339828491, | |
| "learning_rate": 1.0356804114498648e-06, | |
| "loss": 0.5028, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4017677782241864, | |
| "grad_norm": 2.885403871536255, | |
| "learning_rate": 9.9232836408905e-07, | |
| "loss": 0.1797, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6026516673362796, | |
| "grad_norm": 4.714044570922852, | |
| "learning_rate": 9.489763167282352e-07, | |
| "loss": 0.1448, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8035355564483728, | |
| "grad_norm": 3.8532094955444336, | |
| "learning_rate": 9.056242693674206e-07, | |
| "loss": 0.1199, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9636836795859589, | |
| "eval_f1": 0.6468026587210636, | |
| "eval_loss": 0.10681386291980743, | |
| "eval_precision": 0.5823359471729261, | |
| "eval_recall": 0.727319587628866, | |
| "eval_runtime": 20.4746, | |
| "eval_samples_per_second": 208.356, | |
| "eval_steps_per_second": 26.081, | |
| "step": 2489 | |
| }, | |
| { | |
| "epoch": 1.004419445560466, | |
| "grad_norm": 1.4627267122268677, | |
| "learning_rate": 8.622722220066057e-07, | |
| "loss": 0.1126, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.2053033346725592, | |
| "grad_norm": 2.071267604827881, | |
| "learning_rate": 8.189201746457909e-07, | |
| "loss": 0.1053, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.4061872237846524, | |
| "grad_norm": 6.656711101531982, | |
| "learning_rate": 7.755681272849762e-07, | |
| "loss": 0.0984, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.6070711128967456, | |
| "grad_norm": 1.1916735172271729, | |
| "learning_rate": 7.322160799241614e-07, | |
| "loss": 0.0966, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.8079550020088389, | |
| "grad_norm": 2.3483071327209473, | |
| "learning_rate": 6.888640325633467e-07, | |
| "loss": 0.0914, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9708767672041989, | |
| "eval_f1": 0.7143198661247907, | |
| "eval_loss": 0.0914178416132927, | |
| "eval_precision": 0.6660722246990638, | |
| "eval_recall": 0.7701030927835052, | |
| "eval_runtime": 19.9542, | |
| "eval_samples_per_second": 213.79, | |
| "eval_steps_per_second": 26.761, | |
| "step": 4978 | |
| }, | |
| { | |
| "epoch": 2.008838891120932, | |
| "grad_norm": 4.1778669357299805, | |
| "learning_rate": 6.455119852025319e-07, | |
| "loss": 0.0846, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.2097227802330255, | |
| "grad_norm": 2.798360586166382, | |
| "learning_rate": 6.02159937841717e-07, | |
| "loss": 0.0857, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.4106066693451185, | |
| "grad_norm": 1.7000788450241089, | |
| "learning_rate": 5.588078904809024e-07, | |
| "loss": 0.0809, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.611490558457212, | |
| "grad_norm": 2.7977373600006104, | |
| "learning_rate": 5.154558431200876e-07, | |
| "loss": 0.083, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.812374447569305, | |
| "grad_norm": 2.01316499710083, | |
| "learning_rate": 4.721037957592728e-07, | |
| "loss": 0.0776, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9727700697378616, | |
| "eval_f1": 0.7365162519347542, | |
| "eval_loss": 0.08494957536458969, | |
| "eval_precision": 0.6844434609426865, | |
| "eval_recall": 0.7971649484536083, | |
| "eval_runtime": 19.9143, | |
| "eval_samples_per_second": 214.218, | |
| "eval_steps_per_second": 26.815, | |
| "step": 7467 | |
| }, | |
| { | |
| "epoch": 3.0132583366813983, | |
| "grad_norm": 0.729656457901001, | |
| "learning_rate": 4.287517483984581e-07, | |
| "loss": 0.0834, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.2141422257934913, | |
| "grad_norm": 4.330155849456787, | |
| "learning_rate": 3.853997010376433e-07, | |
| "loss": 0.0772, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.4150261149055847, | |
| "grad_norm": 2.2002146244049072, | |
| "learning_rate": 3.420476536768285e-07, | |
| "loss": 0.075, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.6159100040176777, | |
| "grad_norm": 2.16339111328125, | |
| "learning_rate": 2.9869560631601374e-07, | |
| "loss": 0.0753, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.816793893129771, | |
| "grad_norm": 1.6610848903656006, | |
| "learning_rate": 2.55343558955199e-07, | |
| "loss": 0.0696, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9732963932221231, | |
| "eval_f1": 0.7483106105512746, | |
| "eval_loss": 0.0842830166220665, | |
| "eval_precision": 0.6928649835345774, | |
| "eval_recall": 0.8134020618556701, | |
| "eval_runtime": 21.3601, | |
| "eval_samples_per_second": 199.718, | |
| "eval_steps_per_second": 25.0, | |
| "step": 9956 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 12445, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2474729943855180.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.0790324588106795e-06, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 8, | |
| "seed": 32 | |
| } | |
| } | |