geocoding / tokenizer_config.json
alvarodt's picture
vocab size 32768
bf2958b
raw
history blame contribute delete
343 Bytes
{
"add_prefix_space": false,
"bos_token": "<|BOS|>",
"cls_token": "<|CLS|>",
"eos_token": "<|EOS|>",
"mask_token": "<|MASK|>",
"model_max_length": 1024,
"name_or_path": "gpt2",
"pad_token": "<|PAD|>",
"sep_token": "<|SEP|>",
"special_tokens_map_file": null,
"tokenizer_class": "GPT2Tokenizer",
"unk_token": "<|UNK|>"
}