| """ | |
| ## 词典扩容 | |
| 32000 <pad> | |
| 32001 但 | |
| """ | |
| from transformers import LlamaTokenizer | |
| tokenizer = LlamaTokenizer.from_pretrained("ziqingyang/chinese-llama-2-7b") | |
| tokenizer.comments = "重新设计了新词表（大小：55296），进一步提升了中文字词的覆盖程度" | |
| """ | |
| ## 词典扩容 | |
| 32000 <pad> | |
| 32001 但 | |
| """ | |
| from transformers import LlamaTokenizer | |
| tokenizer = LlamaTokenizer.from_pretrained("ziqingyang/chinese-llama-2-7b") | |
| tokenizer.comments = "重新设计了新词表（大小：55296），进一步提升了中文字词的覆盖程度" | |