gpt-2-10000 / tokenizer_config.json
Baljinnyam's picture
Training in progress, epoch 1
78b6b97
raw
history blame
233 Bytes
{
"add_prefix_space": true,
"bos_token": "<|endoftext|>",
"clean_up_tokenization_spaces": true,
"eos_token": "<|endoftext|>",
"model_max_length": 1024,
"tokenizer_class": "GPT2Tokenizer",
"unk_token": "<|endoftext|>"
}