tinystories-tokenizer / tokenizer_config.json
DmitriyYurckML's picture
Add TinyStories BPE tokenizer (8k vocab)
752493a verified
Raw
History Blame Contribute Delete
200 Bytes
{
"tokenizer_class": "PreTrainedTokenizerFast",
"do_lower_case": false,
"bos_token": "<bos>",
"eos_token": "<eos>",
"pad_token": "<pad>",
"unk_token": "<pad>",
"model_max_length": 2048
}