| { | |
| "best_metric": 0.7624731472235634, | |
| "best_model_checkpoint": "training_dir/checkpoint-6000", | |
| "epoch": 1.7225180081428124, | |
| "eval_steps": 1000, | |
| "global_step": 11000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15659254619480112, | |
| "grad_norm": 67.27922058105469, | |
| "learning_rate": 9.92108585858586e-06, | |
| "loss": 0.7405, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.15659254619480112, | |
| "eval_accuracy": 0.6914335664335665, | |
| "eval_f1": 0.6593208421129121, | |
| "eval_loss": 1.1309396028518677, | |
| "eval_precision": 0.6956839140215929, | |
| "eval_recall": 0.6914335664335665, | |
| "eval_runtime": 47.9571, | |
| "eval_samples_per_second": 47.709, | |
| "eval_steps_per_second": 5.964, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.31318509238960224, | |
| "grad_norm": 14.41289234161377, | |
| "learning_rate": 9.763257575757577e-06, | |
| "loss": 0.4907, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.31318509238960224, | |
| "eval_accuracy": 0.743006993006993, | |
| "eval_f1": 0.733830578366595, | |
| "eval_loss": 0.8223879337310791, | |
| "eval_precision": 0.7463673332921382, | |
| "eval_recall": 0.743006993006993, | |
| "eval_runtime": 47.4569, | |
| "eval_samples_per_second": 48.212, | |
| "eval_steps_per_second": 6.027, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.46977763858440336, | |
| "grad_norm": 1.8675850629806519, | |
| "learning_rate": 9.605429292929293e-06, | |
| "loss": 0.4543, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.46977763858440336, | |
| "eval_accuracy": 0.7312062937062938, | |
| "eval_f1": 0.7152230962884388, | |
| "eval_loss": 0.9456853270530701, | |
| "eval_precision": 0.7333745521313426, | |
| "eval_recall": 0.7312062937062938, | |
| "eval_runtime": 47.3928, | |
| "eval_samples_per_second": 48.277, | |
| "eval_steps_per_second": 6.035, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6263701847792045, | |
| "grad_norm": 36.809181213378906, | |
| "learning_rate": 9.44760101010101e-06, | |
| "loss": 0.4431, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.6263701847792045, | |
| "eval_accuracy": 0.7456293706293706, | |
| "eval_f1": 0.7418425557235627, | |
| "eval_loss": 0.7822393774986267, | |
| "eval_precision": 0.7420175571942784, | |
| "eval_recall": 0.7456293706293706, | |
| "eval_runtime": 47.4108, | |
| "eval_samples_per_second": 48.259, | |
| "eval_steps_per_second": 6.032, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7829627309740056, | |
| "grad_norm": 1.452012300491333, | |
| "learning_rate": 9.289772727272728e-06, | |
| "loss": 0.4423, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.7829627309740056, | |
| "eval_accuracy": 0.7539335664335665, | |
| "eval_f1": 0.7466934407427619, | |
| "eval_loss": 0.8257411122322083, | |
| "eval_precision": 0.7528625679945705, | |
| "eval_recall": 0.7539335664335665, | |
| "eval_runtime": 47.5443, | |
| "eval_samples_per_second": 48.123, | |
| "eval_steps_per_second": 6.015, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9395552771688067, | |
| "grad_norm": 16.726686477661133, | |
| "learning_rate": 9.131944444444445e-06, | |
| "loss": 0.4505, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.9395552771688067, | |
| "eval_accuracy": 0.7670454545454546, | |
| "eval_f1": 0.7624731472235634, | |
| "eval_loss": 0.7416993379592896, | |
| "eval_precision": 0.7669396954578515, | |
| "eval_recall": 0.7670454545454546, | |
| "eval_runtime": 47.5056, | |
| "eval_samples_per_second": 48.163, | |
| "eval_steps_per_second": 6.02, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.096147823363608, | |
| "grad_norm": 47.86187744140625, | |
| "learning_rate": 8.974116161616161e-06, | |
| "loss": 0.4028, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.096147823363608, | |
| "eval_accuracy": 0.7399475524475524, | |
| "eval_f1": 0.7251034362957096, | |
| "eval_loss": 1.0867348909378052, | |
| "eval_precision": 0.7436508232814505, | |
| "eval_recall": 0.7399475524475524, | |
| "eval_runtime": 47.6192, | |
| "eval_samples_per_second": 48.048, | |
| "eval_steps_per_second": 6.006, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.252740369558409, | |
| "grad_norm": 1.5419590473175049, | |
| "learning_rate": 8.816287878787879e-06, | |
| "loss": 0.3891, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.252740369558409, | |
| "eval_accuracy": 0.7504370629370629, | |
| "eval_f1": 0.7360022991843806, | |
| "eval_loss": 1.1146304607391357, | |
| "eval_precision": 0.7541770480812687, | |
| "eval_recall": 0.7504370629370629, | |
| "eval_runtime": 47.6132, | |
| "eval_samples_per_second": 48.054, | |
| "eval_steps_per_second": 6.007, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.4093329157532102, | |
| "grad_norm": 48.14420700073242, | |
| "learning_rate": 8.658459595959596e-06, | |
| "loss": 0.3957, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.4093329157532102, | |
| "eval_accuracy": 0.7613636363636364, | |
| "eval_f1": 0.7534759031290063, | |
| "eval_loss": 0.9045655727386475, | |
| "eval_precision": 0.7581869302928095, | |
| "eval_recall": 0.7613636363636364, | |
| "eval_runtime": 47.5148, | |
| "eval_samples_per_second": 48.153, | |
| "eval_steps_per_second": 6.019, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.5659254619480114, | |
| "grad_norm": 5.539546489715576, | |
| "learning_rate": 8.500631313131314e-06, | |
| "loss": 0.401, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.5659254619480114, | |
| "eval_accuracy": 0.7622377622377622, | |
| "eval_f1": 0.7562280560821415, | |
| "eval_loss": 0.951400876045227, | |
| "eval_precision": 0.760121427774359, | |
| "eval_recall": 0.7622377622377622, | |
| "eval_runtime": 47.6294, | |
| "eval_samples_per_second": 48.038, | |
| "eval_steps_per_second": 6.005, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.7225180081428124, | |
| "grad_norm": 44.26694107055664, | |
| "learning_rate": 8.342803030303031e-06, | |
| "loss": 0.3903, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.7225180081428124, | |
| "eval_accuracy": 0.7552447552447552, | |
| "eval_f1": 0.7447143825904207, | |
| "eval_loss": 0.9456614255905151, | |
| "eval_precision": 0.7531489883387709, | |
| "eval_recall": 0.7552447552447552, | |
| "eval_runtime": 47.5072, | |
| "eval_samples_per_second": 48.161, | |
| "eval_steps_per_second": 6.02, | |
| "step": 11000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 63860, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.2525788225157308e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |