codestar12
/

spe_unigram_1b_50_3k_mpt2mix

spe_unigram_1b_50_3k_mpt2mix / tokenizer_config.json

Upload tokenizer

379cd3e 10 months ago

No virus

254 Bytes

	{
	"bos_token": "<s>",
	"clean_up_tokenization_spaces": false,
	"eos_token": "</s>",
	"model_max_length": 1000000000000000019884624838656,
	"tokenizer_class": "PreTrainedTokenizerFast",
	"unk_token": "<unk>",
	"vocab_file": "mpt2-mix-spe.model"
	}