Upload tokenizer_config.json with huggingface_hub

tokenizer_config.json

@@ -1,4 +1,19 @@
 {
-    "tokenizer_class": "NanoChatTokenizer",
-    "chat_template": "{% if messages[0]['role'] == 'system' %}<|bos|><|user_start|>{{ messages[0]['content'] }}\n\n{{ messages[1]['content'] }}<|user_end|>{% set messages = messages[2:] %}{% else %}<|bos|>{% endif %}{% for message in messages %}{% if loop.index0 % 2 == 0 %}<|user_start|>{{ message['content'] }}<|user_end|>{% else %}<|assistant_start|>{{ message['content'] }}<|assistant_end|>{% endif %}{% endfor %}{% if add_generation_prompt %}<|assistant_start|>{% endif %}"
+    "tokenizer_class": "PreTrainedTokenizerFast",
+    "bos_token": "<|bos|>",
+    "eos_token": "<|assistant_end|>",
+    "pad_token": "<|assistant_end|>",
+    "additional_special_tokens": [
+        "<|user_start|>",
+        "<|user_end|>",
+        "<|assistant_start|>",
+        "<|python_start|>",
+        "<|python_end|>",
+        "<|output_start|>",
+        "<|output_end|>"
+    ],
+    "model_input_names": [
+        "input_ids",
+        "attention_mask"
+    ]
 }