DCLM-1B-v0 / tokenizer_config.json
achal-tri's picture
add tokenizer.json (#1)
c0a1664 verified
raw
history blame
156 Bytes
{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "tokenizer_class": "GPTNeoXTokenizer"}