mix_tok_v2 / tokenizer_config.json
teven's picture
Upload tokenizer
f0817b9
{
"bos_token": "<|startoftext|>",
"eos_token": "<|endoftext|>",
"model_max_length": 1000000000000000019884624838656,
"pad_token": "<|pad|>",
"special_tokens_map_file": "geclm/tokenizers/split_digits/all_data_concatenated/special_tokens_map.json",
"tokenizer_class": "PreTrainedTokenizerFast",
"unk_token": "<|unk|>"
}