Text Generation on Wikipedia Chinese (test)
25.91Lexical DivergenceHuman
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Human2024.10 | 25.91 | 2.93 | |
| Llama-3.1Model Size (# parameters)=8B2024.10 | 33.29 | 10.32 | |
| Mistral-NemoModel Size (# parameters)=12B2024.10 | 34.78 | 12.84 | |
| Llama-3Model Size (# parameters)=8B2024.10 | 36.88 | 15.45 | |
| Qwen2Model Size (# parameters)=7B2024.10 | 37.08 | 20.66 | |
| Mistral-v0.3Model Size (# parameters)=7B2024.10 | 39.02 | 17.29 | |
| Qwen1.5Model Size (# parameters)=7B2024.10 | 41 | 23.33 |