OpenWebMath-tokenizer-1 / tokenizer_config.json
parislo's picture
Upload tokenizer
97c86b4 verified
{
"added_tokens_decoder": {},
"clean_up_tokenization_spaces": true,
"merges_file": "./openwebmath_tokenizer/merges.txt",
"model_max_length": 1000000000000000019884624838656,
"tokenizer_class": "PreTrainedTokenizerFast",
"vocab_file": "./openwebmath_tokenizer/vocab.json"
}