claude-tokenizer / tokenizer_config.json
Xenova's picture
Xenova HF staff
Update tokenizer_config.json (#2)
cae6888 verified
{
"add_prefix_space": false,
"bos_token": "<EOT>",
"clean_up_tokenization_spaces": true,
"eos_token": "<EOT>",
"model_max_length": 200000,
"tokenizer_class": "GPT2TokenizerFast",
"unk_token": "<EOT>"
}