40k-openbook-finetuned-deberta-v3-large-mcqa-TPU-v11
This model is a fine-tuned version of byteprobe/40k-openbook-finetuned-deberta-v3-large-mcqa-TPU-v10 on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 0.5896
- Validation Loss: 0.8204
- Train Map3: 0.8114
- Train Lr: 1.2467371e-11
- Epoch: 0
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'Adam', 'weight_decay': 0.001, 'clipnorm': 1, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': False, 'is_legacy_optimizer': False, 'learning_rate': {'class_name': 'CosineDecay', 'config': {'initial_learning_rate': 2e-06, 'decay_steps': 629, 'alpha': 5e-09, 'name': None}}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
- training_precision: mixed_bfloat16
Training results
| Train Loss | Validation Loss | Train Map3 | Train Lr | Epoch |
|---|---|---|---|---|
| 0.5896 | 0.8204 | 0.8114 | 1.2467371e-11 | 0 |
Framework versions
- Transformers 4.35.0.dev0
- TensorFlow 2.12.0
- Datasets 2.14.5
- Tokenizers 0.14.1