Update tokenizer_config.json
Browse files- tokenizer_config.json +12 -13
tokenizer_config.json
CHANGED
|
@@ -2049,16 +2049,15 @@
|
|
| 2049 |
"special": true
|
| 2050 |
}
|
| 2051 |
},
|
| 2052 |
-
|
| 2053 |
-
|
| 2054 |
-
|
| 2055 |
-
|
| 2056 |
-
|
| 2057 |
-
|
| 2058 |
-
|
| 2059 |
-
|
| 2060 |
-
|
| 2061 |
-
|
| 2062 |
-
|
| 2063 |
-
|
| 2064 |
-
|
|
|
|
| 2049 |
"special": true
|
| 2050 |
}
|
| 2051 |
},
|
| 2052 |
+
"bos_token": "<|begin_of_text|>",
|
| 2053 |
+
"chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + '<|start_header_id|>system<|end_header_id|>\n\n' + 'Akshara-8B: An advanced AI model developed by SVECTOR, tailored for India’s diverse linguistic landscape. I am designed to understand and generate text in multiple Indian languages, including Hindi, Gujarati, Marathi, Tamil, Telugu, Kannada, Punjabi, and English, with a focus on efficiency, accuracy, and scalability. I provide seamless multilingual processing, balancing creativity and precision to support a wide range of applications for users in India and beyond.' + '<|eot_id|>' + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
|
| 2054 |
+
"clean_up_tokenization_spaces": true,
|
| 2055 |
+
"eos_token": "<|end_of_text|>",
|
| 2056 |
+
"model_input_names": [
|
| 2057 |
+
"input_ids",
|
| 2058 |
+
"attention_mask"
|
| 2059 |
+
],
|
| 2060 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 2061 |
+
"pad_token": "<|end_of_text|>",
|
| 2062 |
+
"tokenizer_class": "PreTrainedTokenizerFast"
|
| 2063 |
+
}
|
|
|