latent_finetuning_encoder / tokenizer_config.json
benetraco's picture
Final upload: all components and pipeline
fc36213 verified
{
"add_prefix_space": false,
"added_tokens_decoder": {
"49406": {
"content": "<|startoftext|>",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": true
},
"49407": {
"content": "<|endoftext|>",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": true
},
"49408": {
"content": "SHIFTS",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"49409": {
"content": "VH",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"49410": {
"content": "WMH2017",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
}
},
"bos_token": "<|startoftext|>",
"clean_up_tokenization_spaces": false,
"do_lower_case": true,
"eos_token": "<|endoftext|>",
"errors": "replace",
"extra_special_tokens": {},
"model_max_length": 77,
"pad_token": "<|endoftext|>",
"tokenizer_class": "CLIPTokenizer",
"unk_token": "<|endoftext|>"
}