polish-gpt2-small / tokenizer_config.json
pkedzia's picture
Added first version of polish-gpt2-small model learnt from scratch
3703680
raw
history blame contribute delete
No virus
322 Bytes
{
"add_prefix_space": false,
"additional_special_tokens": [
"bos_token",
"eos_token",
"unk_token"
],
"bos_token": "<|endoftext|>",
"clean_up_tokenization_spaces": true,
"eos_token": "<|endoftext|>",
"model_max_length": 1024,
"tokenizer_class": "GPT2Tokenizer",
"unk_token": "<|endoftext|>"
}