| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 9954, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.10046212577858148, | |
| "grad_norm": 7.991936206817627, | |
| "learning_rate": 1.1042131937295525e-06, | |
| "loss": 0.5537, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.20092425155716295, | |
| "grad_norm": 3.070322036743164, | |
| "learning_rate": 1.0659548196137755e-06, | |
| "loss": 0.2017, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3013863773357444, | |
| "grad_norm": 3.75642728805542, | |
| "learning_rate": 1.0276964454979987e-06, | |
| "loss": 0.1702, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4018485031143259, | |
| "grad_norm": 2.4024105072021484, | |
| "learning_rate": 9.894380713822216e-07, | |
| "loss": 0.132, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5023106288929073, | |
| "grad_norm": 4.069411277770996, | |
| "learning_rate": 9.511796972664448e-07, | |
| "loss": 0.1321, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.6027727546714888, | |
| "grad_norm": 3.244875192642212, | |
| "learning_rate": 9.129213231506679e-07, | |
| "loss": 0.1122, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.7032348804500703, | |
| "grad_norm": 10.270066261291504, | |
| "learning_rate": 8.746629490348911e-07, | |
| "loss": 0.1124, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.8036970062286518, | |
| "grad_norm": 7.5877861976623535, | |
| "learning_rate": 8.364045749191142e-07, | |
| "loss": 0.113, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.9041591320072333, | |
| "grad_norm": 1.5884909629821777, | |
| "learning_rate": 7.981462008033373e-07, | |
| "loss": 0.1058, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9671778827175835, | |
| "eval_f1": 0.6902838679898454, | |
| "eval_loss": 0.10421454906463623, | |
| "eval_precision": 0.6249477643125784, | |
| "eval_recall": 0.7708762886597939, | |
| "eval_runtime": 20.2703, | |
| "eval_samples_per_second": 210.456, | |
| "eval_steps_per_second": 26.344, | |
| "step": 4977 | |
| }, | |
| { | |
| "epoch": 1.0046212577858147, | |
| "grad_norm": 14.664758682250977, | |
| "learning_rate": 7.598878266875604e-07, | |
| "loss": 0.1006, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.1050833835643963, | |
| "grad_norm": 4.180013179779053, | |
| "learning_rate": 7.216294525717837e-07, | |
| "loss": 0.0931, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.2055455093429777, | |
| "grad_norm": 5.196892261505127, | |
| "learning_rate": 6.833710784560068e-07, | |
| "loss": 0.0921, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.306007635121559, | |
| "grad_norm": 1.217821717262268, | |
| "learning_rate": 6.451127043402298e-07, | |
| "loss": 0.0893, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.4064697609001406, | |
| "grad_norm": 3.8249542713165283, | |
| "learning_rate": 6.068543302244529e-07, | |
| "loss": 0.0962, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.5069318866787222, | |
| "grad_norm": 1.3417003154754639, | |
| "learning_rate": 5.68595956108676e-07, | |
| "loss": 0.093, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.6073940124573036, | |
| "grad_norm": 5.087819576263428, | |
| "learning_rate": 5.303375819928991e-07, | |
| "loss": 0.0852, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.707856138235885, | |
| "grad_norm": 4.749046802520752, | |
| "learning_rate": 4.920792078771223e-07, | |
| "loss": 0.0859, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.8083182640144666, | |
| "grad_norm": 5.057666301727295, | |
| "learning_rate": 4.5382083376134543e-07, | |
| "loss": 0.0946, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.908780389793048, | |
| "grad_norm": 2.6310770511627197, | |
| "learning_rate": 4.155624596455685e-07, | |
| "loss": 0.0897, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9725726984312636, | |
| "eval_f1": 0.7473998804542737, | |
| "eval_loss": 0.09381680935621262, | |
| "eval_precision": 0.696989966555184, | |
| "eval_recall": 0.8056701030927835, | |
| "eval_runtime": 19.9896, | |
| "eval_samples_per_second": 213.411, | |
| "eval_steps_per_second": 26.714, | |
| "step": 9954 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 14931, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 1038987366336480.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.1424715678453293e-06, | |
| "num_train_epochs": 3, | |
| "per_device_train_batch_size": 4, | |
| "seed": 8 | |
| } | |
| } | |