Fix Conversational Widget (#16)
Browse files- Fix Conversational Widget (11cfce0f220b0999f7993e22ccf4d24d8ba55577)
Co-authored-by: Lucain Pouget <[email protected]>
- tokenizer_config.json +21 -2
    	
        tokenizer_config.json
    CHANGED
    
    | @@ -1,4 +1,23 @@ | |
| 1 | 
             
            {
         | 
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
| 2 | 
             
              "model_max_length": 1024,
         | 
| 3 | 
            -
              " | 
| 4 | 
            -
             | 
|  | |
|  | 
|  | |
| 1 | 
             
            {
         | 
| 2 | 
            +
              "add_bos_token": false,
         | 
| 3 | 
            +
              "add_prefix_space": false,
         | 
| 4 | 
            +
              "added_tokens_decoder": {
         | 
| 5 | 
            +
                "50256": {
         | 
| 6 | 
            +
                  "content": "<|endoftext|>",
         | 
| 7 | 
            +
                  "lstrip": false,
         | 
| 8 | 
            +
                  "normalized": true,
         | 
| 9 | 
            +
                  "rstrip": false,
         | 
| 10 | 
            +
                  "single_word": false,
         | 
| 11 | 
            +
                  "special": true
         | 
| 12 | 
            +
                }
         | 
| 13 | 
            +
              },
         | 
| 14 | 
            +
              "bos_token": "<|endoftext|>",
         | 
| 15 | 
            +
              "chat_template": "{% for message in messages %}{{ message.content }}{{ eos_token }}{% endfor %}",
         | 
| 16 | 
            +
              "clean_up_tokenization_spaces": true,
         | 
| 17 | 
            +
              "eos_token": "<|endoftext|>",
         | 
| 18 | 
            +
              "errors": "replace",
         | 
| 19 | 
             
              "model_max_length": 1024,
         | 
| 20 | 
            +
              "pad_token": null,
         | 
| 21 | 
            +
              "tokenizer_class": "GPT2Tokenizer",
         | 
| 22 | 
            +
              "unk_token": "<|endoftext|>"
         | 
| 23 | 
            +
            }
         | 

 
		