Update config.json

config.json

@@ -56,16 +56,16 @@
   },
   "vision_config": {
     "model_type": "davit",
-    "drop_path_rate": 0.1,  
+    "drop_path_rate": 0.1,
     "patch_size": [7, 3, 3, 3],  
     "patch_stride": [4, 2, 2, 2],  
     "patch_padding": [3, 1, 1, 1],  
     "patch_prenorm": [false, true, true, true],  
-    "enable_checkpoint": false,  
+    "enable_checkpoint": false,
     "dim_embed": [256, 512, 1024, 2048],  
     "num_heads": [8, 16, 32, 64],  
     "num_groups": [8, 16, 32, 64],  
-    "depths": [1, 1, 9, 1],  
+    "depths": [1, 1, 9, 1],
     "window_size": 12,
     "projection_dim": 1024,
     "visual_temporal_embedding": {
@@ -79,7 +79,6 @@
     "image_feature_source": ["spatial_avg_pool", "temporal_avg_pool"]
   },
   "vocab_size": 51289,
-  "torch_dtype": "float32",
-  "transformers_version": "4.41.0.dev0",
+  "transformers_version": "4.48.3",
   "is_encoder_decoder": true
 }