Upload MllamaForConditionalGeneration

config.json

@@ -19,44 +19,18 @@
       "multi_modal_projector",
       "merger",
       "modality_projection",
-      "vision_model.transformer.layers.0.mlp",
-      "vision_model.transformer.layers.1.mlp",
-      "vision_model.transformer.layers.2.mlp",
-      "vision_model.transformer.layers.3.mlp",
-      "vision_model.transformer.layers.4.mlp",
-      "vision_model.transformer.layers.5.mlp",
-      "vision_model.transformer.layers.6.mlp",
-      "vision_model.transformer.layers.7.mlp",
-      "vision_model.transformer.layers.8.mlp",
-      "vision_model.transformer.layers.9.mlp",
-      "vision_model.transformer.layers.10.mlp",
-      "vision_model.transformer.layers.11.mlp",
-      "vision_model.transformer.layers.12.mlp",
-      "vision_model.transformer.layers.13.mlp",
-      "vision_model.transformer.layers.14.mlp",
-      "vision_model.transformer.layers.15.mlp",
-      "vision_model.transformer.layers.16.mlp",
-      "vision_model.transformer.layers.17.mlp",
-      "vision_model.transformer.layers.18.mlp",
-      "vision_model.transformer.layers.19.mlp",
-      "vision_model.transformer.layers.20.mlp",
-      "vision_model.transformer.layers.21.mlp",
-      "vision_model.transformer.layers.22.mlp",
-      "vision_model.transformer.layers.23.mlp",
-      "vision_model.transformer.layers.24.mlp",
-      "vision_model.transformer.layers.25.mlp",
-      "vision_model.transformer.layers.26.mlp",
-      "vision_model.transformer.layers.27.mlp",
-      "vision_model.transformer.layers.28.mlp",
-      "vision_model.transformer.layers.29.mlp",
-      "vision_model.transformer.layers.30.mlp",
-      "vision_model.transformer.layers.31.mlp",
-      "vision_model.global_transformer.layers.0.mlp",
-      "vision_model.global_transformer.layers.1.mlp",
-      "vision_model.global_transformer.layers.2.mlp",
-      "vision_model.global_transformer.layers.3.mlp",
-      "vision_model.global_transformer.layers.4.mlp",
-      "vision_model.global_transformer.layers.5.mlp"
+      "language_model.model.layers.1.self_attn",
+      "language_model.model.layers.1.mlp",
+      "vision_model.transformer.layers.13.mlp.fc2",
+      "language_model.model.layers.8.cross_attn.o_proj",
+      "language_model.model.layers.13.cross_attn.o_proj",
+      "language_model.model.layers.13.mlp.down_proj",
+      "language_model.model.layers.18.cross_attn.o_proj",
+      "language_model.model.layers.18.mlp.down_proj",
+      "language_model.model.layers.23.cross_attn.o_proj",
+      "language_model.model.layers.28.cross_attn.o_proj",
+      "language_model.model.layers.33.cross_attn.o_proj",
+      "language_model.model.layers.38.cross_attn.o_proj"
     ],
     "llm_int8_threshold": 6.0,
     "load_in_4bit": true,

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:df871653461fc0de46e87c4ee385e4aa992867ca244439ca7ba52798299e64e5
-size 4990770382
+oid sha256:afad208b9b805474e7f5d8de82043cf970f6e799c0b655cd6c6e522b9371e618
+size 4971535505

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cd435129de51a39511eef736db475e561c794a0011910e1f96cf5c5ebd844f8d
-size 2974910239
+oid sha256:fc63401a455ef422970874eabac70ebf6380eb565c592341c4932092f569f720
+size 2937067316