Upload LlavaNextForConditionalGeneration

config.json

@@ -66,7 +66,7 @@
   "multimodal_projector_bias": true,
   "projector_hidden_act": "gelu",
   "text_config": {
-    "_name_or_path": "/mnt/scratch-artemis/gviveiros/TowerVision/llava-next-native/towerp_2b_base/",
+    "_name_or_path": "/mnt/scratch-artemis/gviveiros/TowerVision/llava-next-native/towerp_2b_instruct/",
     "add_faster_video": false,
     "add_time_instruction": false,
     "architectures": [
@@ -76,10 +76,12 @@
     "attention_dropout": 0.0,
     "attn_logit_softcapping": 50.0,
     "cache_implementation": "hybrid",
+    "eos_token_id": 107,
     "faster_token_stride": 10,
     "final_logit_softcapping": 30.0,
     "force_sample": false,
     "head_dim": 256,
+    "hidden_act": "gelu_pytorch_tanh",
     "hidden_activation": "gelu_pytorch_tanh",
     "hidden_size": 2304,
     "image_aspect_ratio": "anyres",
@@ -145,7 +147,7 @@
     "image_split_resolution": null,
     "initializer_range": 0.02,
     "intermediate_size": 9216,
-    "max_position_embeddings": 4096,
+    "max_position_embeddings": 8192,
     "mm_hidden_size": 1152,
     "mm_newline_position": "grid",
     "mm_patch_merge_type": "spatial_unpad",

generation_config.json

@@ -2,7 +2,7 @@
   "_from_model_config": true,
   "bos_token_id": 2,
   "cache_implementation": "hybrid",
-  "eos_token_id": 1,
+  "eos_token_id": 107,
   "pad_token_id": 0,
   "transformers_version": "4.52.0.dev0"
 }

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a337ac8a8db3caab6b1b452762056b731a5a057f3121ab0928244e1d08722246
+oid sha256:a50290cb17c028cfd9e69030c9395071cd80d8ba84faba832bdfc6b0d2149aef
 size 4978392440

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aadce153b8abd872f2e01d65324a98621db80b32209c58df8a7c5ed8cb66de98
+oid sha256:9ab4b06305e64535bf8f944262d881a888b2cff49cbda8dc4e4f0407080ed7c3
 size 1061826432