| language-modelling-on-the-pile | Test-Time Fine-Tuning with SIFT + Llama-3.2 (3B) | #1 | Bits per byte: 0.557 |
| language-modelling-on-the-pile | Test-Time Fine-Tuning with SIFT + Phi-3 (3.8B) | #2 | Bits per byte: 0.595 |
| language-modelling-on-the-pile | Test-Time Fine-Tuning with SIFT + Llama-3.2 (1B) | #3 | Bits per byte: 0.606 |
| language-modelling-on-the-pile | Gemma-2 27B | #4 | Bits per byte: 0.629 |
| language-modelling-on-the-pile | Llama-3.2 3B | #6 | Bits per byte: 0.640 |
| language-modelling-on-the-pile | Phi-3 14B | #8 | Bits per byte: 0.651 |
| language-modelling-on-the-pile | Gemma-2 9B | #9 | Bits per byte: 0.670 |
| language-modelling-on-the-pile | Phi-3 7B | #10 | Bits per byte: 0.678 |
| language-modelling-on-the-pile | Phi-3 3.8B | #11 | Bits per byte: 0.679 |
| language-modelling-on-the-pile | Llama-3.2 1B | #12 | Bits per byte: 0.697 |
| language-modelling-on-the-pile | Gemma-2 2B | #14 | Bits per byte: 0.721 |
| language-modelling-on-the-pile | Llama-3.2-Instruct 3B | #15 | Bits per byte: 0.737 |
| language-modelling-on-the-pile | Test-Time Fine-Tuning with SIFT + GPT-2 (774M) | #17 | Bits per byte: 0.762 |
| language-modelling-on-the-pile | Llama-3.2-Instruct 1B | #19 | Bits per byte: 0.807 |
| language-modelling-on-the-pile | Test-Time Fine-Tuning with SIFT + GPT-2 (124M) | #21 | Bits per byte: 0.862 |