koala-7B-HF / tokenizer_config.json

Update tokenizer_config.json to match other models, like Vicuna - the Koala conversion process does not fill out this file, but based on the code I think these special tokens should be defined.

2d12271 over 1 year ago

raw

history blame

726 Bytes

	{
	"add_bos_token": true,
	"add_eos_token": false,
	"bos_token": {
	"__type": "AddedToken",
	"content": "<s>",
	"lstrip": false,
	"normalized": true,
	"rstrip": false,
	"single_word": false
	},
	"clean_up_tokenization_spaces": false,
	"eos_token": {
	"__type": "AddedToken",
	"content": "</s>",
	"lstrip": false,
	"normalized": true,
	"rstrip": false,
	"single_word": false
	},
	"model_max_length": 1000000000000000019884624838656,
	"pad_token": null,
	"sp_model_kwargs": {},
	"tokenizer_class": "LlamaTokenizer",
	"unk_token": {
	"__type": "AddedToken",
	"content": "<unk>",
	"lstrip": false,
	"normalized": true,
	"rstrip": false,
	"single_word": false
	}
	}