Upload files with `vila-upload`.

Upload tokenizer_config.json
Upload model-00004-of-00007.safetensors
Upload config.json
Upload model-00007-of-00007.safetensors
Upload configuration_vila.py
Upload model-00002-of-00007.safetensors
Upload generation_config.json
Upload model-00005-of-00007.safetensors
Upload special_tokens_map.json
Upload model-00006-of-00007.safetensors
Upload merges.txt
Upload added_tokens.json
Upload model.safetensors.index.json
Upload processing_vila.py
Upload model-00001-of-00007.safetensors
Upload model-00003-of-00007.safetensors
Upload vocab.json
Upload processor_config.json
Upload modeling_vila.py
Upload chat_template.json
Upload preprocessor_config.json

added_tokens.json

@@ -0,0 +1,29 @@
+{
+  "</tool_call>": 151658,
+  "<image>": 151666,
+  "<tool_call>": 151657,
+  "<vila/sentinel>": 151665,
+  "<vila/video>": 151667,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652,
+  "[BOS]": 151668,
+  "[PAD]": 151669
+}

chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<image>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+}

config.json

@@ -0,0 +1,69 @@
+{
+  "architectures": [
+    "VILAForConditionalGeneration"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_vila.VILAConfig",
+    "AutoModel": "modeling_vila.VILAForConditionalGeneration",
+    "AutoModelForCausalLM": "modeling_vila.VILAForConditionalGeneration",
+    "AutoModelForImageTextToText": "modeling_vila.VILAForConditionalGeneration",
+    "AutoModelForVision2Seq": "modeling_vila.VILAForConditionalGeneration"
+  },
+  "hidden_size": 5120,
+  "image_token_id": 151666,
+  "mm_hidden_size": 1152,
+  "mm_projector_type": "mlp_downsample_3x3_fix",
+  "mm_vision_select_feature": "cls_patch",
+  "mm_vision_select_layer": -2,
+  "model_type": "vila",
+  "text_config": {
+    "architectures": [
+      "Qwen2ForCausalLM"
+    ],
+    "attention_dropout": 0.0,
+    "bos_token_id": 151643,
+    "eos_token_id": 151645,
+    "hidden_act": "silu",
+    "hidden_size": 5120,
+    "initializer_range": 0.02,
+    "intermediate_size": 13824,
+    "max_position_embeddings": 32768,
+    "max_window_layers": 70,
+    "model_max_length": 16384,
+    "model_type": "qwen2",
+    "num_attention_heads": 40,
+    "num_hidden_layers": 48,
+    "num_key_value_heads": 8,
+    "rms_norm_eps": 1e-06,
+    "rope_scaling": null,
+    "rope_theta": 1000000.0,
+    "sliding_window": null,
+    "torch_dtype": "bfloat16",
+    "use_cache": true,
+    "use_sliding_window": false,
+    "vocab_size": 151670
+  },
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.50.0",
+  "vision_config": {
+    "architectures": [
+      "SiglipVisionModel"
+    ],
+    "attention_dropout": 0.0,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 1152,
+    "image_size": 448,
+    "intermediate_size": 4304,
+    "layer_norm_eps": 1e-06,
+    "model_type": "siglip_vision_model",
+    "num_attention_heads": 16,
+    "num_channels": 3,
+    "num_hidden_layers": 27,
+    "num_image_tokens": 256,
+    "patch_size": 14,
+    "projection_dim": 2048,
+    "projector_hidden_act": "gelu_fast",
+    "torch_dtype": "bfloat16",
+    "vision_use_head": false
+  }
+}

configuration_vila.py

@@ -0,0 +1,55 @@
+from typing import Any, Dict, Optional
+
+from transformers.configuration_utils import PretrainedConfig
+from transformers.models.qwen2.configuration_qwen2 import Qwen2Config
+from transformers.models.siglip.configuration_siglip import SiglipVisionConfig
+
+
+class VILAConfig(PretrainedConfig):
+    # Class attributes.
+    model_type: str = "vila"
+    sub_configs: Dict[str, PretrainedConfig] = {
+        "text_config": Qwen2Config(),
+        "vision_config": SiglipVisionConfig(),
+    }
+    _auto_class: Optional[str] = "AutoConfig"
+
+    # Configuration for sub-modules.
+    text_config: Qwen2Config = Qwen2Config()
+    vision_config: SiglipVisionConfig = SiglipVisionConfig()
+
+    # Model configuration.
+    hidden_size: int
+    image_token_id: int
+    mm_hidden_size: int
+    mm_projector_type: str
+    mm_vision_select_feature: str
+    mm_vision_select_layer: int
+
+    def __init__(
+        self,
+        *,
+        text_config: Optional[Dict[str, Any]] = None,
+        vision_config: Optional[Dict[str, Any]] = None,
+        hidden_size: Optional[int] = None,
+        image_token_id: Optional[int] = None,
+        mm_hidden_size: Optional[int] = None,
+        mm_projector_type: Optional[str] = None,
+        mm_vision_select_feature: Optional[str] = None,
+        mm_vision_select_layer: Optional[int] = None,
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+
+        self.text_config = Qwen2Config(**text_config) if text_config else Qwen2Config()
+        self.vision_config = SiglipVisionConfig(**vision_config) if vision_config else SiglipVisionConfig()
+
+        # By default, we use settings from NVILA-Lite.
+        self.hidden_size = hidden_size if hidden_size is not None else 1536
+        self.image_token_id = image_token_id if image_token_id is not None else 151649
+        self.mm_hidden_size = mm_hidden_size if mm_hidden_size is not None else 1152
+        self.mm_projector_type = mm_projector_type if mm_projector_type is not None else "mlp_downsample_3x3_fix"
+        self.mm_vision_select_feature = (
+            mm_vision_select_feature if mm_vision_select_feature is not None else "cls_patch"
+        )
+        self.mm_vision_select_layer = mm_vision_select_layer if mm_vision_select_layer is not None else -2

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "pad_token_id": 151643,
+  "transformers_version": "4.50.0"
+}

model-00001-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2d5de32badbcea06b3c48d9db7c17f45f1e57dbe7ecb3a1153c99be63f586e7
+size 4982177040

model-00002-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43e35bbbfe408066278fd06b4093f84ca07496e1ad7071abd514b9e2e3304083
+size 4954847776

model-00003-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad40eba9db9c012b967500fe3f55006af1c0246bce103875544f4b865841bce4
+size 4954847824

model-00004-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0257ca66b82b96c1fb906bf22adc8c68e131db226e5eda5cd1c4c9b3e27dfca
+size 4954847824

model-00005-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02d8102e55af5c5bc4fef3f7824b5ef4bd39bc50009ea99fb9cdd790befb7762
+size 4954847824

model-00006-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e67d849dc9b28e68c55ae917164354558a2d2502c1244b3298495bc3147b97e3
+size 4995856896

model-00007-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fb0c3d9f9175496b72378dd0b25cd153be73d2d1a17aec67acd522354bf9bac
+size 720921232

model.safetensors.index.json

@@ -0,0 +1,1033 @@
+{
+  "metadata": {
+    "total_size": 30518216928
+  },
+  "weight_map": {
+    "llm.lm_head.weight": "model-00006-of-00007.safetensors",
+    "llm.model.embed_tokens.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.0.input_layernorm.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.0.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.0.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.0.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.0.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.0.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.1.input_layernorm.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.1.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.1.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.1.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.1.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.1.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.10.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.10.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.10.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.10.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.10.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.10.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.10.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.10.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.10.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.10.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.10.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.10.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.11.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.11.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.11.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.11.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.11.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.11.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.11.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.11.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.11.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.11.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.11.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.11.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.12.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.12.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.12.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.12.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.12.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.12.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.12.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.12.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.12.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.12.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.12.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.12.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.13.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.13.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.13.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.13.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.13.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.13.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.13.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.13.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.13.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.13.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.13.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.13.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.14.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.14.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.14.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.14.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.14.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.14.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.14.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.14.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.14.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.14.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.14.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.14.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.15.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.15.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.15.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.15.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.15.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.15.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.15.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.15.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.15.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.15.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.15.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.15.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.16.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.16.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.16.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.16.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.16.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.16.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.16.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.16.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.16.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.16.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.16.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.16.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.17.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.17.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.17.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.17.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.17.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.17.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.17.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.17.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.17.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.17.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.17.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.17.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.18.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.18.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.18.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.18.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.18.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.18.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.18.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.18.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.18.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.18.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.18.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.18.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.19.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.19.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.19.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.19.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.19.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.19.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.19.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.19.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.19.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.19.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.19.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.19.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.2.input_layernorm.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.2.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.2.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.2.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.2.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.2.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.20.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.20.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.20.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.20.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.20.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.20.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.20.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.20.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.20.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.20.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.20.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.20.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.21.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.21.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.21.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.21.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.21.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.21.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.21.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.21.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.21.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.21.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.21.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.21.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.22.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.22.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.22.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.22.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.22.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.22.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.22.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.22.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.22.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.22.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.22.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.22.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.23.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.23.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.23.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.23.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.23.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.23.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.23.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.23.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.23.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.23.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.23.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.23.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.24.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.24.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.24.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.24.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.24.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.24.self_attn.k_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.24.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.24.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.24.self_attn.q_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.24.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.24.self_attn.v_proj.bias": "model-00003-of-00007.safetensors",
+    "llm.model.layers.24.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "llm.model.layers.25.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.25.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.25.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.25.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.25.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.25.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.25.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.25.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.25.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.25.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.25.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.25.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.26.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.26.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.26.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.26.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.26.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.26.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.26.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.26.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.26.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.26.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.26.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.26.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.27.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.27.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.27.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.27.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.27.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.27.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.27.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.27.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.27.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.27.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.27.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.27.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.28.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.28.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.28.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.28.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.28.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.28.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.28.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.28.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.28.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.28.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.28.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.28.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.29.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.29.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.29.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.29.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.29.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.29.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.29.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.29.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.29.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.29.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.29.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.29.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.3.input_layernorm.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.3.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.3.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.3.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.3.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.3.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.3.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.30.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.30.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.30.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.30.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.30.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.30.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.30.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.30.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.30.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.30.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.30.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.30.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.31.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.31.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.31.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.31.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.31.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.31.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.31.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.31.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.31.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.31.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.31.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.31.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.32.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.32.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.32.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.32.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.32.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.32.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.32.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.32.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.32.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.32.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.32.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.32.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.33.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.33.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.33.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.33.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.33.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.33.self_attn.k_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.33.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.33.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.33.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.33.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.33.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "llm.model.layers.33.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "llm.model.layers.34.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.34.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.34.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.34.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.34.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.34.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.34.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.34.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.34.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.34.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.34.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.34.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.35.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.35.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.35.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.35.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.35.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.35.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.35.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.35.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.35.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.35.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.35.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.35.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.36.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.36.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.36.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.36.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.36.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.36.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.36.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.36.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.36.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.36.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.36.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.36.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.37.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.37.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.37.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.37.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.37.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.37.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.37.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.37.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.37.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.37.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.37.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.37.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.38.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.38.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.38.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.38.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.38.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.38.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.38.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.38.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.38.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.38.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.38.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.38.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.39.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.39.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.39.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.39.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.39.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.39.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.39.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.39.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.39.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.39.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.39.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.39.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.4.input_layernorm.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.4.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.4.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.4.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.4.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.4.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.4.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.4.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.4.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.4.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.4.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.4.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.40.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.40.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.40.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.40.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.40.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.40.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.40.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.40.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.40.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.40.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.40.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.40.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.41.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.41.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.41.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.41.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.41.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.41.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.41.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.41.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.41.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.41.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.41.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.41.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.42.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.42.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.42.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.42.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.42.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.42.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.42.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.42.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.42.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.42.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.42.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "llm.model.layers.42.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "llm.model.layers.43.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.43.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.43.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.43.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.43.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.43.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.43.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.43.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.43.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.43.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.43.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.43.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.44.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.44.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.44.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.44.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.44.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.44.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.44.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.44.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.44.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.44.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.44.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.44.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.45.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.45.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.45.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.45.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.45.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.45.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.45.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.45.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.45.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.45.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.45.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.45.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.46.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.46.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.46.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.46.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.46.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.46.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.46.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.46.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.46.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.46.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.46.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.46.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.47.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.47.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.47.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.47.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.47.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.47.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.47.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.47.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.47.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.47.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.47.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "llm.model.layers.47.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "llm.model.layers.5.input_layernorm.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.5.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.5.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.5.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.5.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.5.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.5.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.5.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.5.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.5.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.5.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.5.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.6.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.6.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.6.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.6.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.6.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.6.self_attn.k_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.6.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.6.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.6.self_attn.q_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.6.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.6.self_attn.v_proj.bias": "model-00001-of-00007.safetensors",
+    "llm.model.layers.6.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
+    "llm.model.layers.7.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.7.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.7.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.7.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.7.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.7.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.7.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.7.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.7.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.7.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.7.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.7.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.8.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.8.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.8.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.8.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.8.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.8.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.8.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.8.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.8.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.8.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.8.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.8.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.9.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.9.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.9.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.9.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.9.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.9.self_attn.k_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.9.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.9.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.9.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.9.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.layers.9.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "llm.model.layers.9.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "llm.model.norm.weight": "model-00006-of-00007.safetensors",
+    "mm_projector.layers.1.bias": "model-00006-of-00007.safetensors",
+    "mm_projector.layers.1.weight": "model-00006-of-00007.safetensors",
+    "mm_projector.layers.2.bias": "model-00006-of-00007.safetensors",
+    "mm_projector.layers.2.weight": "model-00006-of-00007.safetensors",
+    "mm_projector.layers.4.bias": "model-00006-of-00007.safetensors",
+    "mm_projector.layers.4.weight": "model-00006-of-00007.safetensors",
+    "mm_projector.layers.5.bias": "model-00006-of-00007.safetensors",
+    "mm_projector.layers.5.weight": "model-00006-of-00007.safetensors",
+    "mm_projector.layers.7.bias": "model-00006-of-00007.safetensors",
+    "mm_projector.layers.7.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.embeddings.patch_embedding.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.embeddings.patch_embedding.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.embeddings.position_embedding.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.layer_norm1.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.layer_norm1.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.layer_norm2.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.layer_norm2.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.mlp.fc1.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.mlp.fc1.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.mlp.fc2.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.mlp.fc2.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.out_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.out_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.layer_norm1.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.layer_norm1.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.layer_norm2.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.layer_norm2.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.mlp.fc1.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.mlp.fc1.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.mlp.fc2.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.mlp.fc2.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.out_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.out_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.layer_norm1.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.layer_norm1.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.layer_norm2.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.layer_norm2.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.mlp.fc1.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.mlp.fc1.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.mlp.fc2.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.mlp.fc2.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.out_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.out_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.layer_norm1.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.layer_norm1.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.out_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.out_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.layer_norm1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.layer_norm1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.layer_norm2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.layer_norm2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.mlp.fc1.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.mlp.fc1.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.mlp.fc2.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.mlp.fc2.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.k_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.k_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.out_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.q_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.q_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.v_proj.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.v_proj.weight": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.post_layernorm.bias": "model-00007-of-00007.safetensors",
+    "vision_tower.vision_model.post_layernorm.weight": "model-00007-of-00007.safetensors"
+  }
+}

modeling_vila.py

@@ -0,0 +1,239 @@
+from typing import List, Optional, Type
+
+import torch
+import torch.nn as nn
+from torch import Tensor
+from transformers.configuration_utils import PretrainedConfig
+from transformers.generation.utils import GenerationMixin
+from transformers.modeling_outputs import BaseModelOutputWithPooling, CausalLMOutputWithPast
+from transformers.modeling_utils import PreTrainedModel
+from transformers.models.qwen2.modeling_qwen2 import Qwen2ForCausalLM
+from transformers.models.siglip.modeling_siglip import SiglipVisionModel
+
+from .configuration_vila import VILAConfig
+
+
+class DownSampleBlock(nn.Module):
+    @staticmethod
+    def flat_square(x: Tensor) -> Tensor:
+        n, w, h, c = x.size()
+        if w % 2 == 1:
+            x = torch.concat([x, torch.zeros((n, 1, h, c), dtype=x.dtype).to(x.device)], dim=1).contiguous()
+            n, w, h, c = x.size()
+        if h % 2 == 1:
+            x = torch.concat([x, torch.zeros((n, w, 1, c), dtype=x.dtype).to(x.device)], dim=2).contiguous()
+            n, w, h, c = x.size()
+        x = x.contiguous()
+        x = x.view(n, w, int(h / 2), int(c * 2))
+        x = x.permute(0, 2, 1, 3).contiguous()
+        x = x.view(n, int(h / 2), int(w / 2), int(c * 4))
+        x = x.permute(0, 2, 1, 3).contiguous()
+        return x
+
+    def forward(self, x: Tensor) -> Tensor:
+        vit_embeds = x
+        h = w = int(vit_embeds.shape[1] ** 0.5)
+        vit_embeds = vit_embeds.reshape(vit_embeds.shape[0], h, w, -1)
+        vit_embeds = self.flat_square(vit_embeds)
+        vit_embeds = vit_embeds.reshape(vit_embeds.shape[0], -1, vit_embeds.shape[-1])
+        return vit_embeds
+
+
+class DownSample3x3BlockFix(nn.Module):
+    @staticmethod
+    def flat_square_3x3(x: Tensor) -> Tensor:
+        n, w, h, c = x.size()
+        if w % 3 != 0:
+            x = torch.concat(
+                [
+                    x,
+                    torch.zeros((n, 3 - (w % 3), h, c), device=x.device, dtype=x.dtype),
+                ],
+                dim=1,
+            ).contiguous()
+            n, w, h, c = x.size()
+        x = x.contiguous()
+        if h % 3 != 0:
+            x = torch.concat(
+                [
+                    x,
+                    torch.zeros((n, w, 3 - (h % 3), c), device=x.device, dtype=x.dtype),
+                ],
+                dim=2,
+            ).contiguous()
+            n, w, h, c = x.size()
+        x = x.view(n, w, int(h / 3), int(c * 3))
+        x = x.permute(0, 2, 1, 3).contiguous()
+        x = x.view(n, int(h / 3), int(w / 3), int(c * 9))
+        x = x.permute(0, 2, 1, 3).contiguous()
+        return x
+
+    def forward(self, x: Tensor) -> Tensor:
+        vit_embeds = x
+        h = w = int(vit_embeds.shape[1] ** 0.5)
+        vit_embeds = vit_embeds.reshape(vit_embeds.shape[0], h, w, -1)
+        vit_embeds = self.flat_square_3x3(vit_embeds)
+        vit_embeds = vit_embeds.reshape(vit_embeds.shape[0], -1, vit_embeds.shape[-1])
+        return vit_embeds
+
+
+class MultimodalProjector(nn.Module):
+    layers: nn.Sequential
+
+    def __init__(
+        self,
+        config: VILAConfig,
+        *args,
+        **kwargs,
+    ):
+        super().__init__(*args, **kwargs)
+
+        match config.mm_projector_type:
+            case "linear":
+                self.layers = nn.Sequential(
+                    nn.Linear(config.vision_config.hidden_size, config.hidden_size),
+                )
+            case "mlp_downsample":
+                self.layers = nn.Sequential(
+                    DownSampleBlock(),
+                    nn.LayerNorm(config.mm_hidden_size * 4),
+                    nn.Linear(config.mm_hidden_size * 4, config.hidden_size),
+                    nn.GELU(),
+                    nn.Linear(config.hidden_size, config.hidden_size),
+                )
+            case "mlp_downsample_3x3_fix":
+                self.layers = nn.Sequential(
+                    DownSample3x3BlockFix(),
+                    nn.LayerNorm(config.mm_hidden_size * 9),
+                    nn.Linear(
+                        config.mm_hidden_size * 9,
+                        config.mm_hidden_size * 3,
+                    ),
+                    nn.GELU(),
+                    nn.LayerNorm(config.vision_config.hidden_size * 3),
+                    nn.Linear(config.vision_config.hidden_size * 3, config.hidden_size),
+                    nn.GELU(),
+                    nn.Linear(config.hidden_size, config.hidden_size),
+                )
+            case _:
+                raise NotImplementedError(f"mm_projector_type={config.mm_projector_type} not implemented.")
+
+    def forward(self, x: Tensor) -> Tensor:
+        return self.layers(x)
+
+
+class VILAForConditionalGeneration(PreTrainedModel, GenerationMixin):
+    config_class: Type[PretrainedConfig] = VILAConfig
+    base_model_prefix: str = "llm"
+    _auto_class = "AutoModelForImageTextToText"
+    _no_split_modules: List[str] = ["MultimodalProjector"]
+    _skip_keys_device_placement: List[str] = ["past_key_values"]
+    supports_gradient_checkpointing = True
+    _supports_flash_attn_2: bool = True
+    _supports_sdpa = True
+
+    config: VILAConfig
+
+    llm: Qwen2ForCausalLM
+    mm_projector: MultimodalProjector
+    vision_tower: SiglipVisionModel
+
+    def __init__(
+        self,
+        config: VILAConfig,
+        *args,
+        **kwargs,
+    ):
+        super().__init__(config, *args, **kwargs)
+
+        self.llm = Qwen2ForCausalLM(config.text_config, *args, **kwargs)
+        self.mm_projector = MultimodalProjector(config).to(dtype=self.dtype)
+        self.vision_tower = SiglipVisionModel(config.vision_config, *args, **kwargs)
+
+        self.post_init()
+
+    def forward(
+        self,
+        *,
+        attention_mask: Optional[Tensor] = None,
+        input_ids: Optional[Tensor] = None,
+        inputs_embeds: Optional[Tensor] = None,
+        pixel_values: Optional[Tensor] = None,
+        **kwargs,
+    ) -> CausalLMOutputWithPast:
+        # Vision info is only used for prefilling.
+        if kwargs.get("past_key_values", None) is not None:
+            pixel_values = None
+
+        inputs_embeds = inputs_embeds.to(dtype=self.dtype) if inputs_embeds is not None else None
+        pixel_values = pixel_values.to(dtype=self.dtype) if pixel_values is not None else None
+
+        if inputs_embeds is None:
+            assert input_ids is not None
+
+            inputs_embeds = self._embed(input_ids, pixel_values)
+        else:
+            assert input_ids is None
+            assert pixel_values is None
+
+        outputs = self.llm.__call__(
+            inputs_embeds=inputs_embeds,
+            attention_mask=attention_mask,
+            **kwargs,
+        )
+
+        return outputs
+
+    def get_output_embeddings(self) -> nn.Module:
+        return self.llm.get_output_embeddings()
+
+    def _embed(
+        self,
+        input_ids: Tensor,
+        pixel_values: Optional[Tensor],
+    ) -> Tensor:
+        """Gets the embedding of the input ids and pixel values.
+
+        Args:
+            input_ids: The input ids.
+            pixel_values: The pixel values.
+
+        Returns:
+            The embedding of the input ids and pixel values.
+        """
+
+        image_token_mask = input_ids == self.config.image_token_id
+
+        text_embedding: Tensor = self.llm.get_input_embeddings().__call__(input_ids * ~image_token_mask)
+
+        if pixel_values is None:
+            return text_embedding
+
+        image_features: BaseModelOutputWithPooling = self.vision_tower.__call__(
+            pixel_values.to(dtype=self.vision_tower.dtype),
+            output_hidden_states=True,
+        )
+        assert image_features.hidden_states is not None
+
+        # Select image feature.
+        selected_layer_output = image_features.hidden_states[self.config.mm_vision_select_layer]
+        match self.config.mm_vision_select_feature:
+            case "cls_patch":
+                selected_feature = selected_layer_output
+            case _:
+                raise NotImplementedError(
+                    f"mm_vision_select_feature={self.config.mm_vision_select_feature} not implemented."
+                )
+
+        # TODO: Support dynamic_s2.
+
+        image_embedding: Tensor = self.mm_projector.__call__(selected_feature.to(dtype=self.dtype))
+
+        n_images, n_feature, dim_feature = image_embedding.shape
+        image_embedding = image_embedding.view(n_images * n_feature, dim_feature)
+
+        text_embedding[image_token_mask.to(device=text_embedding.device)] = image_embedding.to(
+            device=text_embedding.device
+        )
+
+        return text_embedding

preprocessor_config.json

@@ -0,0 +1,27 @@
+{
+  "auto_map": {
+    "AutoProcessor": "processing_vila.VILAProcessor"
+  },
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "SiglipImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "processor_class": "VILAProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 448,
+    "width": 448
+  }
+}

processing_vila.py

@@ -0,0 +1,325 @@
+from typing import List, Optional, Tuple, cast
+
+import numpy as np
+import transformers.image_transforms as image_transforms
+import transformers.image_utils as image_utils
+from numpy.typing import NDArray
+from PIL.Image import Image
+from torch import Tensor
+from transformers.feature_extraction_utils import BatchFeature
+from transformers.image_processing_utils import BaseImageProcessor
+from transformers.image_processing_utils_fast import BaseImageProcessorFast
+from transformers.image_utils import ImageInput, VideoInput
+from transformers.models.siglip.image_processing_siglip import SiglipImageProcessor
+from transformers.models.siglip.image_processing_siglip_fast import SiglipImageProcessorFast
+from transformers.processing_utils import ProcessingKwargs, ProcessorMixin, Unpack
+from transformers.tokenization_utils import PreTrainedTokenizer
+from transformers.tokenization_utils_base import PreTrainedTokenizerBase, TextInput
+
+
+class VILAProcessorKwargs(ProcessingKwargs, total=False):
+    _defaults = {}  # type: ignore
+
+
+class VILAProcessorOutput(BatchFeature):
+    input_ids: List[List[int]] | NDArray[np.int64] | Tensor
+    attention_mask: List[List[int]] | NDArray[np.int64] | Tensor
+    pixel_values: Optional[List[NDArray[np.float32]] | NDArray[np.float32] | Tensor]
+
+
+class VILAProcessor(ProcessorMixin):
+    attributes: List[str] = [
+        "image_processor",
+        "tokenizer",
+    ]
+    image_processor_class: str = "AutoImageProcessor"
+    tokenizer_class: str = "AutoTokenizer"
+    _auto_class: str = "AutoProcessor"
+    valid_kwargs: List[str] = [
+        "chat_template",
+        "image_pad_len",
+        "max_tiles",
+        "min_tiles",
+    ]
+
+    # Attributes.
+    image_processor: BaseImageProcessor | BaseImageProcessorFast
+    tokenizer: PreTrainedTokenizerBase
+
+    # Configuration parameters.
+    image_pad_len: int
+    max_tiles: int
+    min_tiles: int
+
+    def __init__(
+        self,
+        image_processor: BaseImageProcessor,
+        tokenizer: PreTrainedTokenizer,
+        *,
+        image_pad_len: Optional[int] = None,
+        max_tiles: Optional[int] = None,
+        min_tiles: Optional[int] = None,
+        **kwargs,
+    ):
+        super().__init__(
+            image_processor,
+            tokenizer,
+            **kwargs,
+        )
+
+        self.image_pad_len = image_pad_len if image_pad_len is not None else 121
+        self.max_tiles = max_tiles if max_tiles is not None else 12
+        self.min_tiles = min_tiles if min_tiles is not None else 1
+
+    def __call__(
+        self,
+        images: Optional[ImageInput] = None,
+        text: Optional[TextInput | List[TextInput]] = None,
+        audio: None = None,
+        videos: Optional[VideoInput] = None,
+        **kwargs: Unpack[VILAProcessorKwargs],
+    ) -> VILAProcessorOutput:
+        # Validate arguments.
+        assert text is not None and text != [], "text must be provided"
+        assert not kwargs.get("is_split_into_words", False), "is_split_into_words=True is not supported"
+
+        output_kwargs = self._merge_kwargs(
+            VILAProcessorKwargs,  # type: ignore
+            tokenizer_init_kwargs=self.tokenizer.init_kwargs,
+            **kwargs,
+        )
+
+        # Process images.
+        if images is not None and images != []:
+            image_inputs, num_cropped_images = self._process_images(
+                images=images,
+                **output_kwargs["images_kwargs"],
+            )
+        else:
+            # If no images are provided, do not define pixel_values.
+            image_inputs = BatchFeature()
+            num_cropped_images = []
+
+        # TODO: video processing.
+
+        # Process text.
+        text = text if isinstance(text, list) else [text]
+
+        text = self._pad_image_tokens_by_num_crops(
+            text,
+            num_cropped_images=num_cropped_images,
+        )
+
+        text = self._pad_image_tokens_by_num_embeddings(
+            text,
+        )
+
+        text_inputs = self.tokenizer.__call__(
+            text,
+            **output_kwargs["text_kwargs"],
+        )
+
+        return VILAProcessorOutput(
+            data={
+                **text_inputs,
+                **image_inputs,
+            }
+        )
+
+    def _crop_image(
+        self,
+        image: Image,
+    ) -> List[Image]:
+        """Crops the image into multiple tiles.
+
+        Args:
+            image: The image to be cropped.
+
+        Returns:
+            The cropped images.
+        """
+
+        # TODO: Support more image processors.
+        assert isinstance(self.image_processor, (SiglipImageProcessor, SiglipImageProcessorFast))
+
+        assert self.image_processor.size["height"] == self.image_processor.size["width"]
+        cropped_size = self.image_processor.size["height"]
+
+        cropped_images: List[Image] = dynamic_preprocess(
+            image,
+            min_num=self.min_tiles,
+            max_num=self.max_tiles,
+            image_size=cropped_size,
+        )
+
+        return cropped_images
+
+    def _pad_image_tokens_by_num_crops(
+        self,
+        text: List[TextInput],
+        *,
+        num_cropped_images: List[int],
+    ) -> List[TextInput]:
+        """Pads each <image> to num_cropped_images of "<image>\n\n".
+
+        Args:
+            text: The text to be padded.
+            num_cropped_images: The number of cropped images for each image token.
+
+        Returns:
+            The padded text.
+        """
+        image_token: str = cast(str, self.tokenizer.image_token)
+
+        # Validate arguments.
+        num_images = len(num_cropped_images)
+        num_image_tokens = sum([item.count(image_token) for item in text])
+        assert num_images == num_image_tokens, (
+            f"Number of image tokens ({num_image_tokens}) in text does not match "
+            f"the number of images ({num_images})."
+        )
+
+        assert all(
+            image_pad_len > 0 for image_pad_len in num_cropped_images
+        ), "All image padding lengths should be positive integers."
+
+        # Pad image tokens.
+        image_idx = 0
+        padded_text: List[TextInput] = []
+
+        for i in range(len(text)):
+            padded_text_item = ""
+            remaining_text = text[i]
+
+            while True:
+                token_pos = remaining_text.find(image_token)
+                if token_pos == -1:
+                    padded_text_item += remaining_text
+                    break
+
+                padded_text_item += remaining_text[:token_pos] + ((image_token + "\n") * num_cropped_images[image_idx])
+
+                image_idx += 1
+                remaining_text = remaining_text[token_pos + len(image_token) :]
+
+            padded_text.append(padded_text_item)
+
+        return padded_text
+
+    def _pad_image_tokens_by_num_embeddings(
+        self,
+        text: List[TextInput],
+    ) -> List[TextInput]:
+        """Pads each <image> to image_pad_len times of "<image>".
+
+        Args:
+            text: The text to be padded.
+
+        Returns:
+            The padded text.
+        """
+        image_token: str = cast(str, self.tokenizer.image_token)
+
+        padded_text: List[TextInput] = []
+
+        for i in range(len(text)):
+            padded_text_item = ""
+            remaining_text = text[i]
+
+            while True:
+                token_pos = remaining_text.find(image_token)
+                if token_pos == -1:
+                    padded_text_item += remaining_text
+                    break
+
+                padded_text_item += remaining_text[:token_pos] + (image_token * self.image_pad_len)
+
+                remaining_text = remaining_text[token_pos + len(image_token) :]
+
+            padded_text.append(padded_text_item)
+
+        return padded_text
+
+    def _process_images(
+        self,
+        images: ImageInput,
+        **kwargs: Unpack[VILAProcessorKwargs],
+    ) -> Tuple[BatchFeature, List[int]]:
+        images_flatten = cast(
+            List[Image] | List[NDArray] | List[Tensor],
+            image_utils.make_flat_list_of_images(images),
+        )
+
+        cropped_images: List[Image] = []
+        num_cropped_images: List[int] = []
+        for image in images_flatten:
+            pil_image: Image = image_transforms.to_pil_image(image)
+            single_cropped_images = self._crop_image(pil_image)
+
+            cropped_images.extend(single_cropped_images)
+            num_cropped_images.append(len(single_cropped_images))
+
+        image_inputs = self.image_processor(
+            cropped_images,
+            **kwargs,
+        )
+
+        return image_inputs, num_cropped_images
+
+
+def dynamic_preprocess(image, min_num=1, max_num=12, image_size=384, use_thumbnail=True):
+    orig_width, orig_height = image.size
+    aspect_ratio = orig_width / orig_height
+
+    # calculate the existing image aspect ratio
+    target_ratios = {
+        (i, j)
+        for n in range(min_num, max_num + 1)
+        for i in range(1, n + 1)
+        for j in range(1, n + 1)
+        if i * j <= max_num and i * j >= min_num
+    }
+    target_ratios = sorted(target_ratios, key=lambda x: x[0] * x[1])
+
+    # find the closest aspect ratio to the target
+    target_aspect_ratio = find_closest_aspect_ratio(aspect_ratio, target_ratios, orig_width, orig_height, image_size)
+
+    # calculate the target width and height
+    target_width = image_size * target_aspect_ratio[0]
+    target_height = image_size * target_aspect_ratio[1]
+    blocks = target_aspect_ratio[0] * target_aspect_ratio[1]
+
+    # resize the image
+    resized_img = image.resize((target_width, target_height))
+    processed_images = []
+    for i in range(blocks):
+        box = (
+            (i % (target_width // image_size)) * image_size,
+            (i // (target_width // image_size)) * image_size,
+            ((i % (target_width // image_size)) + 1) * image_size,
+            ((i // (target_width // image_size)) + 1) * image_size,
+        )
+        # split the image
+        split_img = resized_img.crop(box)
+        processed_images.append(split_img)
+    assert len(processed_images) == blocks
+    if use_thumbnail and len(processed_images) != 1:
+        thumbnail_img = image.resize((image_size, image_size))
+        processed_images.append(thumbnail_img)
+    return processed_images
+
+
+def find_closest_aspect_ratio(aspect_ratio, target_ratios, width, height, image_size):
+    best_ratio_diff = float("inf")
+    best_ratio = (1, 1)
+    area = width * height
+    for ratio in target_ratios:
+        target_aspect_ratio = ratio[0] / ratio[1]
+        ratio_diff = abs(aspect_ratio - target_aspect_ratio)
+        if ratio_diff < best_ratio_diff:
+            best_ratio_diff = ratio_diff
+            best_ratio = ratio
+        elif ratio_diff == best_ratio_diff:
+            if area > 0.5 * image_size * image_size * ratio[0] * ratio[1]:
+                best_ratio = ratio
+    return best_ratio

processor_config.json

@@ -0,0 +1,9 @@
+{
+  "auto_map": {
+    "AutoProcessor": "processing_vila.VILAProcessor"
+  },
+  "image_pad_len": 121,
+  "max_tiles": 12,
+  "min_tiles": 1,
+  "processor_class": "VILAProcessor"
+}

special_tokens_map.json

@@ -0,0 +1,42 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>",
+    "<vila/sentinel>",
+    "<image>",
+    "<vila/video>"
+  ],
+  "bos_token": {
+    "content": "[BOS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image>",
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,260 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<vila/sentinel>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151666": {
+      "content": "<image>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151667": {
+      "content": "<vila/video>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151668": {
+      "content": "[BOS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151669": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>",
+    "<vila/sentinel>",
+    "<image>",
+    "<vila/video>"
+  ],
+  "auto_map": {
+    "AutoProcessor": "processing_vila.VILAProcessor"
+  },
+  "bos_token": "[BOS]",
+  "chat_template": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {
+    "image_token": "<image>"
+  },
+  "image_token": "<image>",
+  "legacy": false,
+  "model_max_length": 16384,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "processor_class": "VILAProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}