tiny-random-z-image / tokenizer /tokenizer_config.json
rkazants's picture
Fix tokenizer for transformers 4.x: add vocab.json/merges.txt, drop 5.x-only config keys (#2)
25878e4
Raw
History Blame Contribute Delete
301 Bytes
{
"add_prefix_space": false,
"bos_token": null,
"clean_up_tokenization_spaces": false,
"eos_token": "<|im_end|>",
"errors": "replace",
"model_max_length": 131072,
"pad_token": "<|endoftext|>",
"split_special_tokens": false,
"tokenizer_class": "Qwen2Tokenizer",
"unk_token": null
}