Language Modelling On The Pile
평가 지표
Bits per byte
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | Bits per byte |
|---|---|
| efficiently-learning-at-test-time-active-fine | 0.557 |
| hungry-hungry-hippos-towards-language | - |
| glm-130b-an-open-bilingual-pre-trained-model | 0.65 |
| efficiently-learning-at-test-time-active-fine | 0.640 |
| efficiently-learning-at-test-time-active-fine | 0.679 |
| the-pile-an-800gb-dataset-of-diverse-text-for | 0.7177 |
| efficiently-learning-at-test-time-active-fine | 0.670 |
| efficiently-learning-at-test-time-active-fine | 0.629 |
| knowledge-unlearning-for-mitigating-privacy | - |
| efficiently-learning-at-test-time-active-fine | 0.721 |
| efficiently-learning-at-test-time-active-fine | 0.862 |
| efficiently-learning-at-test-time-active-fine | 0.678 |
| knowledge-unlearning-for-mitigating-privacy | - |
| hungry-hungry-hippos-towards-language | - |
| specialized-language-models-with-cheap | - |
| knowledge-unlearning-for-mitigating-privacy | - |
| specialized-language-models-with-cheap | - |
| efficiently-learning-at-test-time-active-fine | 0.762 |
| the-pile-an-800gb-dataset-of-diverse-text-for | 1.0928 |
| the-pile-an-800gb-dataset-of-diverse-text-for | 1.0468 |
| efficiently-learning-at-test-time-active-fine | 0.595 |
| efficiently-learning-at-test-time-active-fine | 0.697 |
| test-time-training-on-nearest-neighbors-for | 0.85 |
| efficiently-learning-at-test-time-active-fine | 0.651 |
| specialized-language-models-with-cheap | - |
| knowledge-unlearning-for-mitigating-privacy | - |
| efficiently-learning-at-test-time-active-fine | 0.807 |
| the-pile-an-800gb-dataset-of-diverse-text-for | 1.0828 |
| glm-130b-an-open-bilingual-pre-trained-model | 0.634 |
| knowledge-unlearning-for-mitigating-privacy | - |
| glm-130b-an-open-bilingual-pre-trained-model | 0.742 |
| the-pile-an-800gb-dataset-of-diverse-text-for | 0.7980 |
| efficiently-learning-at-test-time-active-fine | 0.737 |
| specialized-language-models-with-cheap | - |
| the-pile-an-800gb-dataset-of-diverse-text-for | 1.2253 |
| the-pile-an-800gb-dataset-of-diverse-text-for | 0.9631 |
| the-pile-an-800gb-dataset-of-diverse-text-for | 0.8718 |
| knowledge-unlearning-for-mitigating-privacy | - |
| efficiently-learning-at-test-time-active-fine | 0.606 |