22 lines
477 B
JSON
22 lines
477 B
JSON
|
{
|
||
|
|
"add_prefix_space": false,
|
||
|
|
"backend": "tokenizers",
|
||
|
|
"bos_token": "<|endoftext|>",
|
||
|
|
"eos_token": "<|endoftext|>",
|
||
|
|
"errors": "replace",
|
||
|
|
"extra_special_tokens": [
|
||
|
|
"<CHANNEL>",
|
||
|
|
"<TITLE>",
|
||
|
|
"<COMMENT>"
|
||
|
|
],
|
||
|
|
"is_local": true,
|
||
|
|
"max_length": 256,
|
||
|
|
"model_max_length": 1024,
|
||
|
|
"pad_token": "<PAD>",
|
||
|
|
"stride": 0,
|
||
|
|
"tokenizer_class": "GPT2Tokenizer",
|
||
|
|
"truncation_side": "right",
|
||
|
|
"truncation_strategy": "longest_first",
|
||
|
|
"unk_token": "<|endoftext|>"
|
||
|
|
}
|