Upload DualTowerVLM using push_to_hub

config.json

@@ -25,8 +25,8 @@
     "lm_max_length": 2048,
     "lm_use_tokens": false,
     "lm_tie_weights": true,
-    "lm_model_type": "HuggingFaceTB/SmolLM2-135M",
-    "lm_tokenizer": "HuggingFaceTB/SmolLM2-360M-Instruct",
+    "lm_model_type": "HuggingFaceTB/SmolLM2-135M-Instruct",
+    "lm_tokenizer": "HuggingFaceTB/SmolLM2-135M-Instruct",
     "lm_chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
     "mp_pixel_shuffle_factor": 4,
     "mp_image_token_length": 64,
@@ -101,13 +101,13 @@
         "r8c8": "<row_8_col_8>"
     },
     "vlm_load_backbone_weights": true,
-    "vlm_checkpoint_path": "checkpoints",
+    "vlm_checkpoint_path": "lusxvr/nanoVLM-230M-8k",
     "hf_repo_name": "nanoVLM",
-    "left_tower_mask_mode": "full",
-    "left_tower_prefill_no_grad": true,
-    "kv_bridge_enabled": true,
+    "left_mask_scope": "full",
+    "use_kv_bridge": true,
     "kv_bridge_type": "mlp",
     "kv_bridge_mlp_ratio": 4.0,
     "kv_bridge_use_rmsnorm": true,
-    "kv_bridge_residual": true
+    "kv_bridge_residual": true,
+    "kv_bridge_init_mode": "default"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:69349a6d14f6ae6f7e90a10f9c824ec6a0df546cffdbfb6cc6f11075ecc74478
+oid sha256:ab55f80ab94f64913dbfcdd026ec8b326f4983663320464c7a7c3f53267489fe
 size 1458447488