Upload config.json with huggingface_hub

config.json

@@ -1,7 +1,7 @@
 {
-  "model_type": "vision_encoder_decoder",
+  "model_type": "llama2",
   "architectures": [
-    "VisionEncoderDecoderModel"
+    "LlamaForVisionText2Text"
   ],
   "_name_or_path": "meta-llama/Llama-3.2-11B-Vision-Instruct",
   "torch_dtype": "bfloat16",
@@ -9,8 +9,17 @@
   "use_cache": true,
   "tie_word_embeddings": false,
   "vision_config": {
-    "model_type": "vision_encoder_decoder",
+    "hidden_size": 1024,
     "image_size": 224,
-    "patch_size": 16
+    "patch_size": 14,
+    "num_hidden_layers": 24,
+    "num_attention_heads": 16
+  },
+  "text_config": {
+    "hidden_size": 4096,
+    "intermediate_size": 11008,
+    "num_hidden_layers": 32,
+    "num_attention_heads": 32,
+    "vocab_size": 32000
   }
 }