mega-ar-small-4096-sw_minipile / tokenizer_config.json
pszemraj's picture
Training in progress, step 100
03896ad
raw
history blame
259 Bytes
{
"add_prefix_space": false,
"bos_token": "<|endoftext|>",
"clean_up_tokenization_spaces": true,
"eos_token": "<|endoftext|>",
"max_length": 4096,
"model_max_length": 4096,
"tokenizer_class": "GPTNeoXTokenizer",
"unk_token": "<|endoftext|>"
}