initial commit

chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{%- if tools is iterable and tools | length > 0 %}\n    {{- '<|im_start|><|system|>'}}\n    {{- '\ub2f9\uc2e0\uc740 \ub3c4\uad6c \ud638\ucd9c \uae30\ub2a5\uc744 \uac16\ucd98 \uc720\uc6a9\ud55c \ub3c4\uc6b0\ubbf8\uc785\ub2c8\ub2e4. \uc0ac\uc6a9\uc790\uc758 \uc694\uccad\uc744 \ucc98\ub9ac\ud558\uae30 \uc704\ud574\uc11c \ud544\uc694\ud55c \ub3c4\uad6c\uac00 \uc8fc\uc5b4\uc9c4 \ubaa9\ub85d\uc5d0 \uc788\ub294 \uacbd\uc6b0 \ub3c4\uad6c \ud638\ucd9c\ub85c \uc751\ub2f5\ud558\uc138\uc694.\n\ud544\uc694\ud55c \ub3c4\uad6c\uac00 \ubaa9\ub85d\uc5d0 \uc5c6\ub294 \uacbd\uc6b0\uc5d0\ub294 \ub3c4\uad6c \ud638\ucd9c \uc5c6\uc774 \uc0ac\uc6a9\uc790\uac00 \uc694\uad6c\ud55c \uc815\ubcf4\ub97c \uc81c\uacf5\ud558\uc138\uc694.\n\ud544\uc694\ud55c \ub3c4\uad6c\uac00 \ubaa9\ub85d\uc5d0 \uc788\uc9c0\ub9cc \ud574\ub2f9 \ub3c4\uad6c\ub97c \ud638\ucd9c\ud558\ub294\ub370 \ud544\uc694\ud55c argument \uc815\ubcf4\uac00 \ubd80\uc871\ud55c \uacbd\uc6b0 \ud574\ub2f9 \uc815\ubcf4\ub97c \uc0ac\uc6a9\uc790\uc5d0\uac8c \uc694\uccad\ud558\uc138\uc694.\n\uc0ac\uc6a9\uc790\uc758 \uc694\uccad\uc744 \ucc98\ub9ac\ud558\uae30 \uc704\ud574 \uc5ec\ub7ec\ubc88 \ub3c4\uad6c\ub97c \ud638\ucd9c\ud560 \uc218 \uc788\uc5b4\uc57c \ud569\ub2c8\ub2e4.\n\ub3c4\uad6c \ud638\ucd9c \uc774\ud6c4 \ub3c4\uad6c \uc2e4\ud589 \uacb0\uacfc\ub97c \uc785\ub825\uc73c\ub85c \ubc1b\uc73c\uba74 \ud574\ub2f9 \uacb0\uacfc\ub97c \ud65c\uc6a9\ud558\uc5ec \ub2f5\ubcc0\uc744 \uc0dd\uc131\ud558\uc138\uc694.\n\n\ub2e4\uc74c\uc740 \uc811\uadfc\ud560 \uc218 \uc788\ub294 \ub3c4\uad6c\ub4e4\uc758 \ubaa9\ub85d \uc785\ub2c8\ub2e4:\n<tools>\n'}}\n    {%- for t in tools %}\n        {{- t | tojson }}\n        {{- '\n' }}\n    {%- endfor %}\n    {{- '</tools>' }}\n    {{- '\n\n\ub3c4\uad6c\ub97c \ud638\ucd9c\ud558\ub824\uba74 \uc544\ub798\uc758 JSON\uc73c\ub85c \uc751\ub2f5\ud558\uc138\uc694.\n\ub3c4\uad6c \ud638\ucd9c \ud615\uc2dd: <tool_call>{\"name\": \ub3c4\uad6c \uc774\ub984, \"arguments\": dictionary \ud615\ud0dc\uc758 \ub3c4\uad6c \uc778\uc790\uac12}</tool_call>' }}\n    \n    {%- if messages[0].role == 'system' %}\n        {{- '\n\n' + messages[0].content}}\n        {% set dummy = messages.pop(0) %}\n    {%- endif %}  \n    {{- '<|im_end|>' }}\n{%- endif %}\n\n{%- for message in messages %}\n    {%- if loop.first and message.role != 'system' %}\n        {{- '<|im_start|><|system|>\ub2f9\uc2e0\uc740 \uc720\uc6a9\ud55c \uc5b8\uc5b4 \ubc0f \uc2dc\uac01 \ub3c4\uc6b0\ubbf8\uc785\ub2c8\ub2e4. \uc0ac\uc6a9\uc790\uac00 \uc81c\uacf5\ud558\ub294 \uc2dc\uac01\uc801 \ucf58\ud150\uce20\ub97c \uc774\ud574\ud560 \uc218 \uc788\uc73c\uba70, \uc790\uc5f0\uc5b4\ub97c \uc0ac\uc6a9\ud558\uc5ec \uc0ac\uc6a9\uc790\uc5d0\uac8c \ub2e4\uc591\ud55c \uc791\uc5c5\uc744 \uc9c0\uc6d0\ud569\ub2c8\ub2e4.<|im_end|>' }}\n    {%- endif %}\n\n    {%- if message.role == 'system' %}\n        {{- '<|im_start|><|system|>' + message.content + '<|im_end|>'}}\n    {%- elif message.role == 'user' %}\n        {%- if message.content is string %}\n            {{- '<|im_start|><|user|>' + message.content + '<|im_end|>'}}\n        {%- else %}\n            {{- '<|im_start|><|user|>' }}\n            {%- for content in message.content %}\n                {%- if content.type == 'image' or 'image' in content or 'image_url' in content %}\n                    {{- '<|extra_id_11|>' }}\n                {%- elif 'text' in content %}\n                    {{- content.text }}\n                {%- endif %}\n            {%- endfor %}\n            {{- '<|im_end|>' }}\n        {%- endif %}\n    {%- elif message.role == 'assistant' %}\n        {{- '<|im_start|><|assistant|>'}}\n        {%- if message.content is defined %}\n            {%- if message.content is string %}\n                {{- message.content }}\n            {%- else %}\n                {%- for content in message.content %}\n                    {%- if content.type == 'image' or 'image' in content or 'image_url' in content %}\n                        {{- '<image>' }}\n                    {%- elif 'text' in content %}\n                        {{- content.text }}\n                    {%- endif %}\n                {%- endfor %}\n            {%- endif %}\n        {%- endif %}\n        {%- if message.tool_calls is defined %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if tool_call.function is defined %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>' }}\n                {{- '{' }}\n                {{- '\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\"' }}\n                {%- if tool_call.arguments is defined %}\n                    {{- ', ' }}\n                    {{- '\"arguments\": ' }}\n                    {{- tool_call.arguments|tojson }}\n                {%- endif %}\n                {{- '}' }}\n                {{- '</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>'}}\n\n    {%- elif message.role == 'tool' %}\n        {{- '<|im_start|><|extra_id_13|><tool_output>' + message.content + '</tool_output><|im_end|>'}}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|><|assistant|>'}}\n{%- endif %}"
+}

config.json

@@ -0,0 +1,152 @@
+{
+  "architectures": [
+    "AX4VLForConditionalGeneration"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_ax4vl.AX4VLConfig",
+    "AutoModelForCausalLM": "modeling_ax4vl.AX4VLForConditionalGeneration",
+    "AutoProcessor": "processing_ax4vl.AX4VLProcessor"
+  },
+  "downsample_ratio": 0.5,
+  "dynamic_image_size": true,
+  "force_image_size": 384,
+  "image_token_index": 22,
+  "llm_config": {
+    "_attn_implementation_autoset": false,
+    "add_cross_attention": false,
+    "architectures": [
+      "Qwen2ForCausalLM"
+    ],
+    "attention_dropout": 0.0,
+    "attn_implementation": "flash_attention_2",
+    "bad_words_ids": null,
+    "begin_suppress_tokens": null,
+    "bos_token_id": 0,
+    "chunk_size_feed_forward": 0,
+    "cross_attention_hidden_size": null,
+    "decoder_start_token_id": null,
+    "diversity_penalty": 0.0,
+    "do_sample": false,
+    "early_stopping": false,
+    "encoder_no_repeat_ngram_size": 0,
+    "eos_token_id": 0,
+    "exponential_decay_length_penalty": null,
+    "finetuning_task": null,
+    "forced_bos_token_id": null,
+    "forced_eos_token_id": null,
+    "hidden_act": "silu",
+    "hidden_size": 3584,
+    "id2label": {
+      "0": "LABEL_0",
+      "1": "LABEL_1"
+    },
+    "initializer_range": 0.02,
+    "intermediate_size": 18944,
+    "is_decoder": false,
+    "is_encoder_decoder": false,
+    "label2id": {
+      "LABEL_0": 0,
+      "LABEL_1": 1
+    },
+    "length_penalty": 1.0,
+    "max_length": 20,
+    "max_position_embeddings": 16384,
+    "max_window_layers": 28,
+    "min_length": 0,
+    "model_type": "qwen2",
+    "no_repeat_ngram_size": 0,
+    "num_attention_heads": 28,
+    "num_beam_groups": 1,
+    "num_beams": 1,
+    "num_hidden_layers": 28,
+    "num_key_value_heads": 4,
+    "num_return_sequences": 1,
+    "output_attentions": false,
+    "output_hidden_states": false,
+    "output_scores": false,
+    "pad_token_id": 1,
+    "prefix": null,
+    "problem_type": null,
+    "pruned_heads": {},
+    "remove_invalid_values": false,
+    "repetition_penalty": 1.0,
+    "return_dict": true,
+    "return_dict_in_generate": false,
+    "rms_norm_eps": 1e-05,
+    "rope_scaling": null,
+    "rope_theta": 1000000,
+    "sep_token_id": null,
+    "sliding_window": null,
+    "suppress_tokens": null,
+    "task_specific_params": null,
+    "temperature": 1.0,
+    "tf_legacy_loss": false,
+    "tie_encoder_decoder": false,
+    "tie_word_embeddings": false,
+    "tokenizer_class": null,
+    "top_k": 50,
+    "top_p": 1.0,
+    "torch_dtype": "bfloat16",
+    "torchscript": false,
+    "typical_p": 1.0,
+    "use_bfloat16": false,
+    "use_cache": false,
+    "use_sliding_window": false,
+    "vocab_size": 102400
+  },
+  "max_dynamic_patch": 12,
+  "max_num_tiles": 12,
+  "min_dynamic_patch": 1,
+  "min_num_tiles": 1,
+  "model_type": "a.x-4-vl",
+  "pad_token_id": 1,
+  "projector_config": {
+    "grid_size": 12,
+    "in_hidden_size": 1152,
+    "model_type": "ldpnetv2_projector",
+    "out_hidden_size": 3584,
+    "torch_dtype": "bfloat16"
+  },
+  "ps_version": "v2",
+  "select_layer": -1,
+  "template": "axvlm",
+  "text_config": {
+    "architectures": [
+      "Qwen2ForCausalLM"
+    ],
+    "attn_implementation": "flash_attention_2",
+    "bos_token_id": 0,
+    "eos_token_id": 0,
+    "hidden_size": 3584,
+    "intermediate_size": 18944,
+    "max_position_embeddings": 16384,
+    "model_type": "qwen2",
+    "num_attention_heads": 28,
+    "num_hidden_layers": 28,
+    "num_key_value_heads": 4,
+    "pad_token_id": 1,
+    "rms_norm_eps": 1e-05,
+    "rope_theta": 1000000,
+    "sliding_window": null,
+    "torch_dtype": "bfloat16",
+    "use_cache": false,
+    "vocab_size": 102400
+  },
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.49.0",
+  "use_thumbnail": true,
+  "vision_config": {
+    "drop_path_rate": 0.0,
+    "hidden_size": 1152,
+    "image_size": 384,
+    "intermediate_size": 4304,
+    "model_type": "siglip_vision_model",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 27,
+    "torch_dtype": "bfloat16",
+    "vision_use_head": false
+  },
+  "vision_feature_layer": 0,
+  "vision_feature_select_strategy": "full"
+}

configuration_ax4vl.py

@@ -0,0 +1,113 @@
+import transformers
+from transformers.utils import logging
+from transformers.models.auto import CONFIG_MAPPING, AutoConfig
+from transformers.configuration_utils import PretrainedConfig
+
+logger = logging.get_logger(__name__)
+
+class LDPConfig(PretrainedConfig):
+    model_type = "ldpnetv2_projector"
+
+    def __init__(
+        self,
+        in_hidden_size=1024,
+        out_hidden_size=2048,
+        grid_size=12,
+        **kwargs
+    ):
+        self.in_hidden_size = in_hidden_size
+        self.out_hidden_size = out_hidden_size
+        self.grid_size = grid_size
+
+        super().__init__(**kwargs)
+
+class MLPProjectorConfig(PretrainedConfig):
+    model_type = "mlp2x_projector"
+
+    def __init__(
+        self,
+        hidden_act="gelu",
+        in_hidden_size=1024,
+        out_hidden_size=2048,
+        bias: bool=True,
+        **kwargs
+    ):
+        self.hidden_act = hidden_act
+        self.in_hidden_size = in_hidden_size
+        self.out_hidden_size = out_hidden_size
+        self.bias = bias
+
+        super().__init__(**kwargs)
+
+
+
+class AX4VLConfig(PretrainedConfig):
+    model_type = "a.x-4-vl"
+    sub_configs = {
+        "text_config": AutoConfig,
+        "projector_config": AutoConfig,
+        "vision_config": AutoConfig
+    }
+
+    def __init__(
+        self,
+        vision_config=None,
+        projector_config=None,
+        text_config=None,
+        image_token_index=102400,
+        vision_feature_select_strategy="full",
+        vision_feature_layer=0,
+        tie_word_embeddings=False,
+        **kwargs,
+    ):
+        self.image_token_index = image_token_index
+
+        if vision_feature_select_strategy not in ["default", "full"]:
+            raise ValueError(
+                "vision_feature_select_strategy should be one of 'default', 'full'."
+                f"Got: {vision_feature_select_strategy}"
+            )
+
+        self.vision_feature_select_strategy = vision_feature_select_strategy
+        self.vision_feature_layer = vision_feature_layer
+
+        if isinstance(vision_config, dict):
+            vision_config["model_type"] = (
+                vision_config["model_type"] if "model_type" in vision_config else "siglip_vision_model"
+            )
+            vision_config = CONFIG_MAPPING[vision_config["model_type"]](**vision_config)
+        elif vision_config is None:
+            vision_config = CONFIG_MAPPING["siglip_vision_model"](
+                intermediate_size=4304,
+                hidden_size=1152,
+                patch_size=16,
+                image_size=384,
+                num_hidden_layers=27,
+                num_attention_heads=16,
+                vision_use_head=False
+            )
+        self.vision_config = vision_config
+
+        if isinstance(projector_config, dict):
+            projector_config["model_type"] = (
+                projector_config["model_type"] if "model_type" in projector_config else "mlp2x"
+            )
+            projector_config = CONFIG_MAPPING[projector_config["model_type"]](**projector_config)
+        elif projector_config is None:
+            projector_config = CONFIG_MAPPING["mlp2x_projector"]()
+        self.projector_config = projector_config
+
+        if isinstance(text_config, dict):
+            text_config["model_type"] = text_config["model_type"] if "model_type" in text_config else "qwen2"
+            text_config = CONFIG_MAPPING[text_config["model_type"]](**text_config)
+        elif text_config is None:
+            text_config = CONFIG_MAPPING["qwen2"]()
+
+        self.text_config = text_config
+
+        super().__init__(tie_word_embeddings=tie_word_embeddings, **kwargs)
+
+
+AutoConfig.register(LDPConfig.model_type, LDPConfig)
+AutoConfig.register(MLPProjectorConfig.model_type, MLPProjectorConfig)
+AutoConfig.register(AX4VLConfig.model_type, AX4VLConfig)

generation_config.json

@@ -0,0 +1,12 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": [
+    0,
+    27,
+    1
+  ],
+  "pad_token_id": 1,
+  "transformers_version": "4.49.0",
+  "use_cache": false
+}

image_processing_ax4vl.py

@@ -0,0 +1,497 @@
+"""
+Image processor class for Megatron-LM LLaVA.
+"""
+
+import math
+from typing import Dict, Iterable, List, Optional, Tuple, Union
+
+import numpy as np
+from PIL import Image
+from .configuration_ax4vl import AX4VLConfig
+
+from transformers.image_utils import (
+    OPENAI_CLIP_MEAN,
+    OPENAI_CLIP_STD,
+    ChannelDimension,
+    ImageInput,
+    PILImageResampling,
+    infer_channel_dimension_format,
+    is_scaled_image,
+    is_valid_image,
+    valid_images,
+    make_list_of_images,
+    to_numpy_array,
+    validate_preprocess_arguments,
+)
+from transformers.image_processing_utils import BatchFeature, get_size_dict, BaseImageProcessor
+from transformers.image_transforms import (
+    PaddingMode,
+    pad,
+    to_channel_dimension_format,
+)
+from transformers.utils import TensorType, logging
+from transformers.models.auto import AutoImageProcessor
+
+
+logger = logging.get_logger(__name__)
+
+def _get_patch_output_size(image, target_resolution):
+    original_width, original_height = image.size
+    target_width, target_height = target_resolution
+
+    scale_w = target_width / original_width
+    scale_h = target_height / original_height
+
+    if scale_w < scale_h:
+        new_width = target_width
+        new_height = min(math.ceil(original_height * scale_w), target_height)
+    else:
+        new_height = target_height
+        new_width = min(math.ceil(original_width * scale_h), target_width)
+
+    return new_width, new_height
+
+# From https://github.com/OpenGVLab/InternVL/blob/c62fa4f7c850165d7386bdc48ac6bc5a6fab0864/internvl_chat/internvl/train/dataset.py#L685
+# Copyright (c) 2023 OpenGVLab.
+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
+    # print(f'width: {width}, height: {height}, best_ratio: {best_ratio}')
+    return best_ratio
+
+def _pad_for_patching(image, target_resolution, background_color=(0, 0, 0)):
+    """
+    Pad an image to a target resolution while maintaining aspect ratio.
+    """
+    target_width, target_height = target_resolution
+    new_width, new_height = _get_patch_output_size(image, target_resolution)
+
+    paste_x = (target_width - new_width) // 2
+    paste_y = (target_height - new_height) // 2
+
+    padded_image = Image.new(image.mode, target_resolution, background_color)
+    padded_image.paste(image, (paste_x, paste_y))
+    return padded_image
+
+def _resize_for_patching(image, target_resolution):
+    new_size = _get_patch_output_size(image, target_resolution)
+
+    # Resize the image
+    resized_image = image.resize(new_size)
+
+    return resized_image
+
+def get_target_ratios(image_size, min_num=1, max_num=6, tile_size=384):
+    orig_width, orig_height = image_size
+    aspect_ratio = orig_width / orig_height
+
+    target_ratios = set(
+        (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])
+
+    return find_closest_aspect_ratio(
+        aspect_ratio, target_ratios, orig_width, orig_height, tile_size
+    )
+
+# From https://github.com/OpenGVLab/InternVL/blob/c62fa4f7c850165d7386bdc48ac6bc5a6fab0864/internvl_chat/internvl/train/dataset.py#L702
+# Copyright (c) 2023 OpenGVLab.
+def dynamic_preprocess(image, min_num=1, max_num=6, image_size=448, use_thumbnail=False, padding=False):
+    # find the closest aspect ratio to the target
+    target_aspect_ratio = get_target_ratios(image.size, min_num=min_num, max_num=max_num, tile_size=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
+    if padding: # LLaVA-Next tiling strategy
+        resized_img = _resize_for_patching(image, (target_width, target_height))
+        resized_img = _pad_for_patching(resized_img, (target_width, target_height))
+    else:   # InternVL tiling strategy
+        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
+
+class AX4VLImageProcessor(BaseImageProcessor):
+
+    model_input_names = ["pixel_values"]
+
+    def __init__(
+        self,
+        do_resize: bool = True,
+        size: Dict[str, int] = None,
+        resample: PILImageResampling = PILImageResampling.BICUBIC,
+        do_rescale: bool = True,
+        rescale_factor: Union[int, float] = 1 / 255,
+        do_normalize: bool = True,
+        image_mean: Optional[Union[float, List[float]]] = None,
+        image_std: Optional[Union[float, List[float]]] = None,
+        do_pad: Optional[bool] = True,
+        do_tile_pad: Optional[bool] = True,
+        do_convert_rgb: bool = True,
+        use_thumbnail: bool = True,
+        min_num_tiles: int = 1,
+        max_num_tiles: int = 6,
+        **kwargs,
+    ) -> None:
+        super().__init__(**kwargs)
+        size = dict(size) if size is not None else {"shortest_edge": 224}
+        size = get_size_dict(size, default_to_square=False)
+
+        self.do_resize = do_resize
+        self.size = size
+        self.resample = resample
+        self.do_rescale = do_rescale
+        self.rescale_factor = rescale_factor
+        self.do_normalize = do_normalize
+        self.image_mean = image_mean if image_mean is not None else OPENAI_CLIP_MEAN
+        self.image_std = image_std if image_std is not None else OPENAI_CLIP_STD
+        self.do_pad = do_pad
+        self.do_tile_pad = do_tile_pad
+        self.do_convert_rgb = do_convert_rgb
+        self.use_thumbnail = use_thumbnail
+        self.min_num_tiles = min_num_tiles
+        self.max_num_tiles = max_num_tiles
+
+    def pad(
+        self,
+        image: np.ndarray,
+        padding: Union[int, Tuple[int, int], Iterable[Tuple[int, int]]],
+        mode: PaddingMode = PaddingMode.CONSTANT,
+        constant_values: Union[float, Iterable[float]] = 0.0,
+        data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
+    ) -> np.ndarray:
+        """
+        Pads the `image` with the specified `padding` and `mode`. Padding can be in the (`height`, `width`)
+        dimension of in the (`num_patches`) dimension. In the second case an iterable if tuples is expected
+        as input.
+
+        Args:
+            image (`np.ndarray`):
+                The image to pad.
+            padding (`int` or `Tuple[int, int]` or `Iterable[Tuple[int, int]]`):
+                Padding to apply to the edges of the height, width axes. Can be one of three formats:
+                - `((before_height, after_height), (before_width, after_width))` unique pad widths for each axis.
+                - `((before, after),)` yields same before and after pad for height and width.
+                - `(pad,)` or int is a shortcut for before = after = pad width for all axes.
+            mode (`PaddingMode`):
+                The padding mode to use. Can be one of:
+                    - `"constant"`: pads with a constant value.
+                    - `"reflect"`: pads with the reflection of the vector mirrored on the first and last values of the
+                    vector along each axis.
+                    - `"replicate"`: pads with the replication of the last value on the edge of the array along each axis.
+                    - `"symmetric"`: pads with the reflection of the vector mirrored along the edge of the array.
+            constant_values (`float` or `Iterable[float]`, *optional*):
+                The value to use for the padding if `mode` is `"constant"`.
+            data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format for the output image. Can be one of:
+                    - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                    - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                If unset, will use same as the input image.
+            input_data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format for the input image. Can be one of:
+                    - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                    - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                If unset, will use the inferred format of the input image.
+
+        Returns:
+            `np.ndarray`: The padded image.
+
+        """
+
+        # call the general `pad` if padding on `height/width`, otherwise it's the `num_patched` dim
+        if isinstance(padding, int) or len(padding) != 4:
+            return pad(image, padding, mode, constant_values, data_format, input_data_format)
+
+        if input_data_format is None:
+            input_data_format = infer_channel_dimension_format(image)
+        if mode == PaddingMode.CONSTANT:
+            image = np.pad(image, padding, mode="constant", constant_values=constant_values)
+        elif mode == PaddingMode.REFLECT:
+            image = np.pad(image, padding, mode="reflect")
+        elif mode == PaddingMode.REPLICATE:
+            image = np.pad(image, padding, mode="edge")
+        elif mode == PaddingMode.SYMMETRIC:
+            image = np.pad(image, padding, mode="symmetric")
+        else:
+            raise ValueError(f"Invalid padding mode: {mode}")
+        image = (
+            to_channel_dimension_format(image, data_format, input_data_format) if data_format is not None else image
+        )
+        return image
+
+    def _pad_for_batching(
+        self,
+        pixel_values: List[np.ndarray],
+        data_format: Optional[Union[str, ChannelDimension]] = None,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
+    ):
+        """
+        Pads images on the `num_of_patches` dimension with zeros to form a batch of same number of patches.
+
+        Args:
+            pixel_values (`List[np.ndarray]`):
+                An array of pixel values of each images of shape (`batch_size`, `num_patches`, `image_in_3D`)
+            data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format for the output image. Can be one of:
+                    - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                    - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                If unset, will use same as the input image.
+            input_data_format (`str` or `ChannelDimension`, *optional*):
+                The channel dimension format for the input image. Can be one of:
+                    - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                    - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                If unset, will use the inferred format of the input image.
+
+        Returns:
+            List[`np.ndarray`]: The padded images.
+        """
+        max_patch = max(len(x) for x in pixel_values)
+        pixel_values = [
+            self.pad(
+                image,
+                padding=((0, max_patch - image.shape[0]), (0, 0), (0, 0), (0, 0)),
+                data_format=data_format,
+                input_data_format=input_data_format,
+            )
+            for image in pixel_values
+        ]
+
+        return pixel_values
+
+    def _preprocess(
+        self,
+        images: ImageInput,
+        do_resize: bool = None,
+        size: Dict[str, int] = None,
+        resample: PILImageResampling = None,
+        do_rescale: bool = None,
+        rescale_factor: float = None,
+        do_normalize: bool = None,
+        image_mean: Optional[Union[float, List[float]]] = None,
+        image_std: Optional[Union[float, List[float]]] = None,
+        do_convert_rgb: bool = None,
+        data_format: Optional[ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
+    ):
+        images = make_list_of_images(images)
+
+        all_images = []
+        for image in images:
+            if do_resize:
+                image = image.resize((size["shortest_edge"], size["shortest_edge"]), resample)
+
+            image = to_numpy_array(image)
+
+            if input_data_format is None:
+                # We assume that all images have the same channel dimension format.
+                input_data_format = infer_channel_dimension_format(image)
+
+            if is_scaled_image(image) and do_rescale:
+                logger.warning_once(
+                    "It looks like you are trying to rescale already rescaled images. If the input"
+                    " images have pixel values between 0 and 1, set `do_rescale=False` to avoid rescaling them again."
+                )
+            if do_rescale:
+                image = self.rescale(image=image, scale=rescale_factor, input_data_format=input_data_format)
+
+            if do_normalize:
+                image = self.normalize(
+                    image=image, mean=image_mean, std=image_std, input_data_format=input_data_format
+                )
+
+            all_images.append(image)
+
+        images = [
+            to_channel_dimension_format(image, data_format, input_channel_dim=input_data_format)
+            for image in all_images
+        ]
+
+        return images
+
+    def preprocess(
+        self,
+        images: ImageInput,
+        do_resize: bool = None,
+        size: Dict[str, int] = None,
+        resample: PILImageResampling = None,
+        do_rescale: bool = None,
+        rescale_factor: float = None,
+        do_normalize: bool = None,
+        image_mean: Optional[Union[float, List[float]]] = None,
+        image_std: Optional[Union[float, List[float]]] = None,
+        do_pad: Optional[bool] = None,
+        do_convert_rgb: bool = None,
+        return_tensors: Optional[Union[str, TensorType]] = None,
+        data_format: Optional[ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
+    ):
+        """
+        Args:
+            images (`ImageInput`):
+                Image to preprocess. Expects a single or batch of images with pixel values ranging from 0 to 255.
+            do_resize (`bool`, *optional*, defaults to `self.do_resize`):
+                Whether to resize the image.
+            size (`Dict[str, int]`, *optional*, defaults to `self.size`):
+                Size of the image after resizing. Shortest edge of the image is resized to size["shortest_edge"], with
+                the longest edge resized to keep the input aspect ratio.
+            resample (`int`, *optional*, defaults to `self.resample`):
+                Resampling filter to use if resizing the image. This can be one of the enum `PILImageResampling`. Only
+                has an effect if `do_resize` is set to `True`.
+            do_normalize (`bool`, *optional*, defaults to `self.do_normalize`):
+                Whether to normalize the image.
+            image_mean (`float` or `List[float]`, *optional*, defaults to `self.image_mean`):
+                Image mean to use for normalization. Only has an effect if `do_normalize` is set to `True`.
+            image_std (`float` or `List[float]`, *optional*, defaults to `self.image_std`):
+                Image standard deviation to use for normalization. Only has an effect if `do_normalize` is set to
+                `True`.
+            do_pad (`bool`, *optional*, defaults to `self.do_pad`):
+                Whether to pad the image. If `True`, will pad the patch dimension of the images in the batch to the largest
+                number of patches in the batch. Padding will be applied to the bottom and right with zeros.
+            do_convert_rgb (`bool`, *optional*, defaults to `self.do_convert_rgb`):
+                Whether to convert the image to RGB.
+            return_tensors (`str` or `TensorType`, *optional*):
+                The type of tensors to return. Can be one of:
+                - Unset: Return a list of `np.ndarray`.
+                - `TensorType.TENSORFLOW` or `'tf'`: Return a batch of type `tf.Tensor`.
+                - `TensorType.PYTORCH` or `'pt'`: Return a batch of type `torch.Tensor`.
+                - `TensorType.NUMPY` or `'np'`: Return a batch of type `np.ndarray`.
+                - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
+            data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
+                The channel dimension format for the output image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
+
+        """
+        do_resize = do_resize if do_resize is not None else self.do_resize
+        size = size if size is not None else self.size
+        size = get_size_dict(size, param_name="size", default_to_square=False)
+        resample = resample if resample is not None else self.resample
+        do_rescale = do_rescale if do_rescale is not None else self.do_rescale
+        rescale_factor = rescale_factor if rescale_factor is not None else self.rescale_factor
+        do_normalize = do_normalize if do_normalize is not None else self.do_normalize
+        image_mean = image_mean if image_mean is not None else self.image_mean
+        image_std = image_std if image_std is not None else self.image_std
+        do_pad = do_pad if do_pad is not None else self.do_pad
+        do_convert_rgb = do_convert_rgb if do_convert_rgb is not None else self.do_convert_rgb
+
+        images = make_batched_images(images)
+
+        if not valid_images(images):
+            raise ValueError(
+                "Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, "
+                "torch.Tensor, tf.Tensor or jax.ndarray."
+            )
+
+        validate_preprocess_arguments(
+            do_rescale=do_rescale,
+            rescale_factor=rescale_factor,
+            do_normalize=do_normalize,
+            image_mean=image_mean,
+            image_std=image_std,
+            do_resize=do_resize,
+            size=size,
+            resample=resample,
+        )
+
+        new_images, num_tiles = [], []
+        image_sizes = [image.size for image in images]
+        for image in images:
+            if do_convert_rgb and image.mode != "RGB":
+                image = image.convert("RGB")
+
+            image_patches = dynamic_preprocess(
+                image,
+                min_num=self.min_num_tiles,
+                max_num=self.max_num_tiles,
+                image_size=self.size["shortest_edge"],
+                use_thumbnail=self.use_thumbnail,
+                padding=self.do_tile_pad
+            )
+
+            # preprocess patches
+            pixel_values = self._preprocess(
+                image_patches,
+                do_resize=do_resize,
+                size=size,
+                resample=resample,
+                do_rescale=do_rescale,
+                rescale_factor=rescale_factor,
+                do_normalize=do_normalize,
+                image_mean=image_mean,
+                image_std=image_std,
+                data_format=data_format,
+                input_data_format=input_data_format
+            )
+            pixel_values = np.array(pixel_values)
+            new_images.append(pixel_values)
+            num_tiles.append(len(image_patches))
+
+        if do_pad:
+            processed_images = self._pad_for_batching(new_images)
+        else:
+            processed_images = np.concatenate(new_images)
+
+        return BatchFeature(
+            data={"pixel_values": processed_images, "image_sizes": image_sizes, "num_tiles": num_tiles},
+            tensor_type=return_tensors
+        )
+
+
+def make_batched_images(images) -> List[List[ImageInput]]:
+    """
+    Accepts images in list or nested list format, and makes a list of images for preprocessing.
+
+    Args:
+        images (`Union[List[List[ImageInput]], List[ImageInput], ImageInput]`):
+            The input image.
+
+    Returns:
+        list: A list of images.
+    """
+    if isinstance(images, (list, tuple)) and isinstance(images[0], (list, tuple)) and is_valid_image(images[0][0]):
+        return [img for img_list in images for img in img_list]
+
+    elif isinstance(images, (list, tuple)) and is_valid_image(images[0]):
+        return images
+
+    elif is_valid_image(images):
+        return [images]
+
+    raise ValueError(f"Could not make batched video from {images}")
+
+AutoImageProcessor.register(AX4VLConfig, AX4VLImageProcessor)

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df2cb01ffd10ce08ecc0bc65bf7574b9b6307255d08223f1f9bb293be670f354
+size 4915685072

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33edd933eba6633abf140bb94f5e2f1b6632cf936a790f58400d07ec2df8b602
+size 4932752832

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d092fc7551ec532351529e9e9f7df911e5748bfbe4083b17114848ddae75194
+size 4796984024

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6263e4575c902f940b9c52a36144a748ba46d92ae0cbfca5689d110594184372
+size 734003344

model.safetensors.index.json

@@ -0,0 +1,789 @@
+{
+  "metadata": {
+    "total_size": 15379320288
+  },
+  "weight_map": {
+    "language_model.lm_head.weight": "model-00004-of-00004.safetensors",
+    "language_model.model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.0.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.0.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.0.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.1.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.1.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.1.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.10.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.10.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.10.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.11.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.11.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.11.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.12.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.12.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.12.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.13.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.13.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.13.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.14.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.14.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.14.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.15.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.15.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.15.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.16.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.16.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.16.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.17.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.17.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.17.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.17.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.17.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.17.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.18.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.18.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.18.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.18.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.18.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.18.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.18.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.18.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.18.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.19.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.19.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.19.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.2.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.2.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.2.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.20.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.20.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.20.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.20.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.21.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.21.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.21.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.22.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.22.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.22.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.23.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.23.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.23.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.24.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.24.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.24.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.25.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.25.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.25.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.26.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.26.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.26.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.27.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.27.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.27.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "language_model.model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.3.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.3.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.3.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.4.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.4.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.4.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.5.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.5.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.5.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.6.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.6.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.6.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.7.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.7.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.7.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.7.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.7.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "language_model.model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.8.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.8.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.8.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.8.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.8.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.8.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.8.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.9.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.9.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.9.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "language_model.model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "language_model.model.norm.weight": "model-00003-of-00004.safetensors",
+    "multi_modal_projector.mlp.mlp.0.bias": "model-00001-of-00004.safetensors",
+    "multi_modal_projector.mlp.mlp.0.weight": "model-00001-of-00004.safetensors",
+    "multi_modal_projector.mlp.mlp.2.bias": "model-00001-of-00004.safetensors",
+    "multi_modal_projector.mlp.mlp.2.weight": "model-00001-of-00004.safetensors",
+    "multi_modal_projector.peg.peg.0.bias": "model-00001-of-00004.safetensors",
+    "multi_modal_projector.peg.peg.0.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.embeddings.patch_embedding.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.embeddings.patch_embedding.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.embeddings.position_embedding.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.24.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.25.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.26.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.layer_norm1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.layer_norm1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.layer_norm2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.layer_norm2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.mlp.fc1.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.mlp.fc1.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.mlp.fc2.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.mlp.fc2.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.out_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.post_layernorm.bias": "model-00001-of-00004.safetensors",
+    "vision_tower.vision_model.post_layernorm.weight": "model-00001-of-00004.safetensors"
+  }
+}

modeling_ax4vl.py

@@ -0,0 +1,385 @@
+"""
+base code: LLaVA-Next (transformers==4.49.0)
+"""
+from typing import List, Optional, Tuple, Union
+import math
+
+import torch
+import torch.utils.checkpoint
+from torch import nn
+from .configuration_ax4vl import LDPConfig, MLPProjectorConfig, AX4VLConfig
+
+from transformers.activations import ACT2FN
+from transformers.generation import GenerationMixin
+from transformers.models.auto import AutoModel, AutoModelForCausalLM
+from transformers.utils import (
+    is_torchdynamo_compiling,
+    logging,
+)
+from transformers.configuration_utils import PretrainedConfig
+from transformers.modeling_utils import PreTrainedModel
+from transformers.modeling_outputs import ModelOutput
+from dataclasses import dataclass
+
+
+
+logger = logging.get_logger(__name__)
+
+def build_projector(config):
+    if config.model_type == "ldpnetv2_projector":
+        return LDPProjector(config)
+    else:
+        raise ValueError(f"Unknown projector type: {config.model_type}")
+
+@dataclass
+class AX4CausalLMOutputWithPast(ModelOutput):
+    loss: Optional[torch.FloatTensor] = None
+    logits: torch.FloatTensor = None
+    past_key_values: Optional[List[torch.FloatTensor]] = None
+    hidden_states: Optional[Tuple[torch.FloatTensor]] = None
+    attentions: Optional[Tuple[torch.FloatTensor]] = None
+    image_hidden_states: Optional[torch.FloatTensor] = None
+
+
+class BaseAXPretrainedModel(PreTrainedModel):
+    config_class = PretrainedConfig
+    base_model_prefix = "model"
+    supports_gradient_checkpointing = True
+    _no_split_modules = ["AXVisionAttention"]
+    _skip_keys_device_placement = "past_key_values"
+    _supports_cache_class = True
+    _supports_flash_attn_2 = True
+    _supports_sdpa = True
+    _supports_quantized_cache = True
+    _supports_static_cache = True
+
+    def __init__(self, config: PretrainedConfig):
+        super().__init__(config)
+
+    def _init_weights(self, module):
+        # important: this ported version of LlavaNext isn't meant for training from scratch - only
+        # inference and fine-tuning - so the proper init weights code has been removed - the original codebase
+        # https://github.com/haotian-liu/LLaVA/tree/main/llava_next should serve for that purpose
+        std = (
+            self.config.initializer_range
+            if hasattr(self.config, "initializer_range")
+            else self.config.text_config.initializer_range
+        )
+
+        if hasattr(module, "class_embedding"):
+            module.class_embedding.data.normal_(mean=0.0, std=std)
+
+        if isinstance(module, (nn.Linear, nn.Conv2d)):
+            module.weight.data.normal_(mean=0.0, std=std)
+            if module.bias is not None:
+                module.bias.data.zero_()
+        elif isinstance(module, nn.Embedding):
+            module.weight.data.normal_(mean=0.0, std=std)
+            if module.padding_idx is not None:
+                module.weight.data[module.padding_idx].zero_()
+
+
+
+class AX4VLForConditionalGeneration(BaseAXPretrainedModel, GenerationMixin):
+    config_class = AX4VLConfig
+
+    def __init__(self, config: AX4VLConfig):
+        super().__init__(config)
+        self.vision_tower = AutoModel.from_config(config.vision_config)
+
+        self.multi_modal_projector = build_projector(config.projector_config)
+        self.vocab_size = config.text_config.vocab_size
+        self.language_model = AutoModelForCausalLM.from_config(config.text_config)
+        if self.language_model._tied_weights_keys is not None:
+            self._tied_weights_keys = [f"language_model.{k}" for k in self.language_model._tied_weights_keys]
+
+        self.pad_token_id = self.config.pad_token_id if self.config.pad_token_id is not None else -1
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.language_model.get_input_embeddings()
+
+    def set_input_embeddings(self, value):
+        self.language_model.set_input_embeddings(value)
+
+    def get_output_embeddings(self):
+        return self.language_model.get_output_embeddings()
+
+    def set_output_embeddings(self, new_embeddings):
+        self.language_model.set_output_embeddings(new_embeddings)
+
+    def set_decoder(self, decoder):
+        self.language_model.set_decoder(decoder)
+
+    def get_decoder(self):
+        return self.language_model.get_decoder()
+
+    def get_image_features(
+        self,
+        pixel_values: torch.FloatTensor,
+        vision_feature_layer: Union[int, List[int]],
+        vision_feature_select_strategy: str,
+    ):
+        if pixel_values.dim() != 4:
+            # otherwise has to be stacked from list of (num_patches, num_channels, height, width)
+            raise ValueError(f"pixel_values of shape {pixel_values.shape}, expect to be of 4 or 5 dimensions")
+
+        image_outputs = self.vision_tower(pixel_values, output_hidden_states=True)
+        # If we have one vision feature layer, return the corresponding hidden states,
+        # otherwise, select the hidden states of each feature layer and concatenate them
+        if isinstance(vision_feature_layer, int):
+            if vision_feature_layer == 0:
+                selected_image_feature = image_outputs.last_hidden_state
+            else:
+                selected_image_feature = image_outputs.hidden_states[vision_feature_layer]
+        else:
+            hs_pool = [image_features.hidden_states[layer_idx] for layer_idx in vision_feature_layer]
+            selected_image_feature = torch.cat(hs_pool, dim=-1)
+
+        if vision_feature_select_strategy == "default":
+            selected_image_feature = selected_image_feature[:, 1:]
+        elif vision_feature_select_strategy == "full":
+            selected_image_feature = selected_image_feature
+
+        image_features = self.multi_modal_projector(selected_image_feature)
+        return image_features
+
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        pixel_values: torch.FloatTensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        vision_feature_layer: Optional[Union[int, List[int]]] = None,
+        vision_feature_select_strategy: Optional[str] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        cache_position: Optional[torch.LongTensor] = None,
+        logits_to_keep: Union[int, torch.Tensor] = 0,
+        **lm_kwargs,
+    ) -> Union[Tuple, AX4CausalLMOutputWithPast]:
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+        vision_feature_layer = (
+            vision_feature_layer if vision_feature_layer is not None else self.config.vision_feature_layer
+        )
+        vision_feature_select_strategy = (
+            vision_feature_select_strategy
+            if vision_feature_select_strategy is not None
+            else self.config.vision_feature_select_strategy
+        )
+
+        if (input_ids is None) ^ (inputs_embeds is not None):
+            raise ValueError("You must specify exactly one of input_ids or inputs_embeds")
+
+        if pixel_values is not None and inputs_embeds is not None:
+            raise ValueError(
+                "You cannot specify both pixel_values and inputs_embeds at the same time, and must specify either one"
+            )
+
+        if inputs_embeds is None:
+            inputs_embeds = self.get_input_embeddings()(input_ids)
+
+        if pixel_values is not None and pixel_values.size(0) > 0:
+            image_features = self.get_image_features(
+                pixel_values,
+                vision_feature_layer=vision_feature_layer,
+                vision_feature_select_strategy=vision_feature_select_strategy,
+            )
+
+            special_image_mask = (input_ids == self.config.image_token_index).unsqueeze(-1)
+            special_image_mask = special_image_mask.expand_as(inputs_embeds).to(inputs_embeds.device)
+            if not is_torchdynamo_compiling() and inputs_embeds[special_image_mask].numel() != image_features.numel():
+                n_image_tokens = (input_ids == self.config.image_token_index).sum()
+                n_image_features = image_features.shape[0]
+                raise ValueError(
+                    f"Image features and image tokens do not match: tokens: {n_image_tokens}, features {n_image_features}"
+                )
+            image_features = image_features.to(inputs_embeds.device, inputs_embeds.dtype)
+            inputs_embeds = inputs_embeds.masked_scatter(special_image_mask, image_features)
+
+        outputs = self.language_model(
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+            cache_position=cache_position,
+            logits_to_keep=logits_to_keep,
+            **lm_kwargs,
+        )
+
+        logits = outputs[0]
+
+        loss = None
+        if labels is not None:
+            # Shift so that tokens < n predict n
+            if attention_mask is not None:
+                # we use the input attention mask to shift the logits and labels, because it is 2D.
+                # we also crop attn mask in case it is longer, which happens in PrefixTuning with peft
+                shift_attention_mask = attention_mask[:, -(logits.shape[1] - 1) :].to(logits.device)
+                shift_logits = logits[..., :-1, :][shift_attention_mask.to(logits.device) != 0].contiguous()
+                shift_labels = labels[..., 1:][shift_attention_mask.to(labels.device) != 0].contiguous()
+            else:
+                shift_logits = logits[..., :-1, :].contiguous()
+                shift_labels = labels[..., 1:].contiguous()
+            # Flatten the tokens
+            loss_fct = nn.CrossEntropyLoss()
+            loss = loss_fct(
+                shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1).to(shift_logits.device)
+            )
+
+        if not return_dict:
+            output = (logits,) + outputs[1:]
+            return (loss,) + output if loss is not None else output
+
+        return AX4CausalLMOutputWithPast(
+            loss=loss,
+            logits=logits,
+            past_key_values=outputs.past_key_values,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+            image_hidden_states=image_features if pixel_values is not None else None,
+        )
+
+    def prepare_inputs_for_generation(
+        self,
+        input_ids,
+        past_key_values=None,
+        inputs_embeds=None,
+        pixel_values=None,
+        image_sizes=None,
+        attention_mask=None,
+        cache_position=None,
+        logits_to_keep=None,
+        **kwargs,
+    ):
+        # Overwritten -- in specific circumstances we don't want to forward image inputs to the model
+
+        model_inputs = self.language_model.prepare_inputs_for_generation(
+            input_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            attention_mask=attention_mask,
+            cache_position=cache_position,
+            logits_to_keep=logits_to_keep,
+            **kwargs,
+        )
+
+        # If we're in cached decoding stage, pixel values should be None because input ids do not contain special image token anymore
+        # Otherwise we need pixel values to be passed to model
+        if cache_position[0] == 0:
+            model_inputs["pixel_values"] = pixel_values
+            model_inputs["image_sizes"] = image_sizes
+
+        return model_inputs
+
+
+
+
+
+class FeatureIRLayer(nn.Module):
+    def __init__(self, in_dim: int, out_dim: int) -> None:
+        super().__init__()
+        self.mlp = nn.Sequential(
+            nn.Linear(in_dim, out_dim), nn.GELU(), nn.Linear(out_dim, out_dim)
+        )
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.mlp(x)
+
+
+class TokenDownLayer(nn.Module):
+    def __init__(self, shape) -> None:
+        super().__init__()
+        self.dwn = nn.Sequential(
+            nn.AdaptiveAvgPool2d(shape)
+        )
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        b, num_tokens, c = x.shape
+        h = int(math.sqrt(num_tokens))
+        assert h * h == num_tokens
+        x = x.permute(0, 2, 1).reshape(b, -1, h, h)
+        x = self.dwn(x)
+        x = x.flatten(2).transpose(1, 2)
+        return x
+    
+
+class PosInjectLayer(nn.Module):
+    # https://github.com/Meituan-AutoML/Twins/blob/main/gvt.py
+    def __init__(
+            self,
+            in_dim: int,
+            out_dim: int,
+            stride: int = 1,
+            padding: int = 1,
+            shape = None) -> None:
+        super().__init__()
+        self.peg = nn.Sequential(
+            nn.Conv2d(in_dim, out_dim, 3, stride, padding, bias=True, groups=out_dim)
+        )
+        self.pool = None
+        if shape is not None:
+            self.pool = nn.Sequential(
+                nn.AdaptiveAvgPool2d(shape)
+            )
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        b, num_tokens, c = x.shape
+        h = int(math.sqrt(num_tokens))
+        assert h * h == num_tokens
+        cnn_feat = x.transpose(1, 2).view(b, c, h, h)
+        if self.pool is not None:
+            x = self.peg(cnn_feat) + self.pool(cnn_feat)
+        else:
+            x = self.peg(cnn_feat) + cnn_feat
+        x = x.flatten(2).transpose(1, 2)
+        return x
+
+class LDPProjector(PreTrainedModel):
+    config_class = LDPConfig
+    _no_split_modules = []
+
+    def __init__(self, config):
+        super().__init__(config)
+        inc, ouc = config.in_hidden_size, config.out_hidden_size
+        grid = config.grid_size
+        self.mlp = FeatureIRLayer(inc, ouc)
+        self.dwn = TokenDownLayer((grid, grid))
+        self.peg = PosInjectLayer(ouc, ouc, stride=1)
+
+    def forward(self, x):
+        x = self.mlp(x)
+        x = self.dwn(x)
+        x = self.peg(x)
+        return x
+
+class MLPProjector(PreTrainedModel):
+    config_class = MLPProjectorConfig
+    _no_split_modules = []
+
+    def __init__(self, config):
+        super().__init__(config)
+
+        self.linear_1 = nn.Linear(config.in_hidden_size, config.out_hidden_size, bias=config.bias)
+        self.act = ACT2FN[config.hidden_act]
+        self.linear_2 = nn.Linear(config.out_hidden_size, config.out_hidden_size, bias=config.bias)
+
+    def forward(self, image_features):
+        hidden_states = self.linear_1(image_features)
+        hidden_states = self.act(hidden_states)
+        hidden_states = self.linear_2(hidden_states)
+        return hidden_states
+

preprocessor_config.json

@@ -0,0 +1,32 @@
+{
+  "auto_map": {
+  "AutoImageProcessor": "image_processing_ax4vl.AX4VLImageProcessor",
+  "AutoProcessor": "processing_ax4vl.AX4VLProcessor"
+  },
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_pad": false,
+  "do_rescale": true,
+  "do_resize": true,
+  "do_tile_pad": false,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "AX4VLImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "max_num_tiles": 12,
+  "min_num_tiles": 1,
+  "processor_class": "AX4VLProcessor",
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "shortest_edge": 384
+  },
+  "use_thumbnail": true
+}

processing_ax4vl.py

@@ -0,0 +1,121 @@
+from typing import List, Union
+from .configuration_ax4vl import AX4VLConfig
+from transformers.models.auto import AutoProcessor
+from transformers.feature_extraction_utils import BatchFeature
+from transformers.image_utils import ImageInput
+from transformers.tokenization_utils_base import PreTokenizedInput, TextInput
+from transformers.processing_utils import ProcessingKwargs, ProcessorMixin, _validate_images_text_input_order
+
+
+
+class BaseAXProcessor(ProcessorMixin):
+    attributes = ["image_processor", "tokenizer"]
+    image_processor_class = "AutoImageProcessor"
+    tokenizer_class = "AutoTokenizer"
+
+
+class AX4VLProcessorKwargs(ProcessingKwargs, total=False):
+    _defaults = {
+        "text_kwargs": {
+            "padding": False,
+        },
+        "images_kwargs": {
+            "do_pad": False,
+        },
+    }
+
+
+class AX4VLProcessor(BaseAXProcessor):
+    valid_kwargs = [
+        "chat_template",
+        "patch_size",
+        "num_tokens_per_tile",
+        "image_token",
+    ]
+
+    def __init__(
+        self,
+        image_processor=None,
+        tokenizer=None,
+        patch_size=16,
+        num_tokens_per_tile=144,
+        image_token="<image>",  # set the default and let users change if they have peculiar special tokens in rare cases
+        chat_template=None,
+        **kwargs
+    ):
+        self.patch_size = patch_size
+        self.num_tokens_per_tile = num_tokens_per_tile
+        self.image_token = tokenizer.image_token if hasattr(tokenizer, "image_token") else image_token
+        super().__init__(image_processor, tokenizer, chat_template=chat_template)
+
+    def __call__(
+        self,
+        images: ImageInput = None,
+        text: Union[TextInput, PreTokenizedInput, List[TextInput], List[PreTokenizedInput]] = None,
+        conversations: List = None,
+        **kwargs
+    ) -> BatchFeature:
+        if images is None and conversations is None and text is None:
+            raise ValueError("You have to specify at least images, text or conversation.")
+
+        if not text and conversations is not None:
+            if isinstance(conversations[0], dict):
+                conversations = [conversations]
+            text = [self.apply_chat_template(conv, **kwargs) for conv in conversations]
+
+        images, text = _validate_images_text_input_order(images, text)
+        
+        output_kwargs = self._merge_kwargs(
+            AX4VLProcessorKwargs,
+            tokenizer_init_kwargs=self.tokenizer.init_kwargs,
+            **kwargs,
+        )
+
+        if images is not None:
+            image_inputs = self.image_processor(images, **output_kwargs["images_kwargs"])
+        else:
+            image_inputs = {}
+
+        prompt_strings = text
+        if image_inputs:
+            num_tiles = iter(image_inputs["num_tiles"])
+            prompt_strings = []
+            for sample in text:
+                while self.image_token in sample:
+                    num_tile = next(num_tiles)
+                    num_image_tokens = num_tile * self.num_tokens_per_tile
+                    sample = sample.replace(self.image_token, "<placeholder>" * num_image_tokens, 1)
+                prompt_strings.append(sample)
+            prompt_strings = [sample.replace("<placeholder>", self.image_token) for sample in prompt_strings]
+
+        text_inputs = self.tokenizer(prompt_strings, **output_kwargs["text_kwargs"])
+        
+        if "num_tiles" in image_inputs:
+            del image_inputs["num_tiles"]
+        return BatchFeature(data={**text_inputs, **image_inputs})
+
+    # Copied from transformers.models.clip.processing_clip.CLIPProcessor.batch_decode with CLIP->Llama
+    def batch_decode(self, *args, **kwargs):
+        """
+        This method forwards all its arguments to LlamaTokenizerFast's [`~PreTrainedTokenizer.batch_decode`]. Please
+        refer to the docstring of this method for more information.
+        """
+        return self.tokenizer.batch_decode(*args, **kwargs)
+
+    # Copied from transformers.models.clip.processing_clip.CLIPProcessor.decode with CLIP->Llama
+    def decode(self, *args, **kwargs):
+        """
+        This method forwards all its arguments to LlamaTokenizerFast's [`~PreTrainedTokenizer.decode`]. Please refer to
+        the docstring of this method for more information.
+        """
+        return self.tokenizer.decode(*args, **kwargs)
+
+    @property
+    # Copied from transformers.models.clip.processing_clip.CLIPProcessor.model_input_names
+    def model_input_names(self):
+        tokenizer_input_names = self.tokenizer.model_input_names
+        image_processor_input_names = self.image_processor.model_input_names
+        return list(dict.fromkeys(tokenizer_input_names + image_processor_input_names))
+
+
+AutoProcessor.register(AX4VLConfig, AX4VLProcessor)

processor_config.json

@@ -0,0 +1,9 @@
+{
+  "auto_map": {
+  "AutoProcessor":  "processing_ax4vl.AX4VLProcessor"
+  },
+  "image_token": "<|extra_id_11|>",
+  "num_tokens_per_tile": 144,
+  "patch_size": 16,
+  "processor_class": "AX4VLProcessor"
+}

special_tokens_map.json

@@ -0,0 +1,90 @@
+{
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<|pad|>",
+    "<|unk|>",
+    "<|sep|>",
+    "<|mask|>",
+    "<|cls|>",
+    "<|image|>",
+    "<|audio|>",
+    "<|user|>",
+    "<|system|>",
+    "<|assistant|>",
+    "<|extra_id_0|>",
+    "<|extra_id_1|>",
+    "<|extra_id_2|>",
+    "<|extra_id_3|>",
+    "<|extra_id_4|>",
+    "<|extra_id_5|>",
+    "<|extra_id_6|>",
+    "<|extra_id_7|>",
+    "<|extra_id_8|>",
+    "<|extra_id_9|>",
+    "<|extra_id_10|>",
+    "<|extra_id_11|>",
+    "<|extra_id_12|>",
+    "<|extra_id_13|>",
+    "<|im_start|>",
+    "<|im_sep|>",
+    "<|im_end|>",
+    "<|resident_reg|>",
+    "<|foreigner_reg|>",
+    "<|business_reg|>",
+    "<|credit_card|>",
+    "<|passport|>",
+    "<|driver_license|>",
+    "<|telephone|>",
+    "<|health_insurance|>",
+    "<|bank_account|>"
+  ],
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<|cls|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<|mask|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|pad|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "<|sep|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|unk|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,395 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<|pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<|unk|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<|sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<|mask|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<|cls|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "<|image|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "<|audio|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "<|system|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "<|extra_id_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "<|extra_id_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "<|extra_id_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<|extra_id_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "15": {
+      "content": "<|extra_id_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "16": {
+      "content": "<|extra_id_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "17": {
+      "content": "<|extra_id_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "18": {
+      "content": "<|extra_id_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "19": {
+      "content": "<|extra_id_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "20": {
+      "content": "<|extra_id_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "21": {
+      "content": "<|extra_id_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "22": {
+      "content": "<|extra_id_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "23": {
+      "content": "<|extra_id_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "24": {
+      "content": "<|extra_id_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "25": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "26": {
+      "content": "<|im_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "27": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "28": {
+      "content": "<|resident_reg|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "29": {
+      "content": "<|foreigner_reg|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30": {
+      "content": "<|business_reg|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "31": {
+      "content": "<|credit_card|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32": {
+      "content": "<|passport|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "33": {
+      "content": "<|driver_license|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "34": {
+      "content": "<|telephone|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "35": {
+      "content": "<|health_insurance|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "36": {
+      "content": "<|bank_account|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "37": {
+      "content": "</tool_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<tool_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<|pad|>",
+    "<|unk|>",
+    "<|sep|>",
+    "<|mask|>",
+    "<|cls|>",
+    "<|image|>",
+    "<|audio|>",
+    "<|user|>",
+    "<|system|>",
+    "<|assistant|>",
+    "<|extra_id_0|>",
+    "<|extra_id_1|>",
+    "<|extra_id_2|>",
+    "<|extra_id_3|>",
+    "<|extra_id_4|>",
+    "<|extra_id_5|>",
+    "<|extra_id_6|>",
+    "<|extra_id_7|>",
+    "<|extra_id_8|>",
+    "<|extra_id_9|>",
+    "<|extra_id_10|>",
+    "<|extra_id_11|>",
+    "<|extra_id_12|>",
+    "<|extra_id_13|>",
+    "<|im_start|>",
+    "<|im_sep|>",
+    "<|im_end|>",
+    "<|resident_reg|>",
+    "<|foreigner_reg|>",
+    "<|business_reg|>",
+    "<|credit_card|>",
+    "<|passport|>",
+    "<|driver_license|>",
+    "<|telephone|>",
+    "<|health_insurance|>",
+    "<|bank_account|>"
+  ],
+  "bos_token": "<|endoftext|>",
+  "chat_template": "{%- if tools is iterable and tools | length > 0 %}\n                {{- '<|im_start|><|system|>'}}\n                {{- '당신은 도구 호출 기능을 갖춘 유용한 도우미입니다. 사용자의 요청을 처리하기 위해서 필요한 도구가 주어진 목록에 있는 경우 도구 호출로 응답하세요.\n필요한 도구가 목록에 없는 경우에는 도구 호출 없이 사용자가 요구한 정보를 제공하세요.\n필요한 도구가 목록에 있지만 해당 도구를 호출하는데 필요한 argument 정보가 부족한 경우 해당 정보를 사용자에게 요청하세요.\n사용자의 요청을 처리하기 위해 여러번 도구를 호출할 수 있어야 합니다.\n도구 호출 이후 도구 실행 결과를 입력으로 받으면 해당 결과를 활용하여 답변을 생성하세요.\n\n다음은 접근할 수 있는 도구들의 목록 입니다:\n<tools>\n'}}\n                {%- for t in tools %}\n                    {{- t | tojson }}\n                    {{- '\n' }}\n                {%- endfor %}\n                {{- '</tools>' }}\n                {{- '\n\n도구를 호출하려면 아래의 JSON으로 응답하세요.\n도구 호출 형식: <tool_call>{\"name\": 도구 이름, \"arguments\": dictionary 형태의 도구 인자값}</tool_call>' }}\n                \n                {%- if messages[0].role == 'system' %}\n                    {{- '\n\n' + messages[0].content}}\n                    {% set dummy = messages.pop(0) %}\n                {%- endif %}  \n                {{- '<|im_end|>' }}\n            {%- endif %}\n    \n            {%- for message in messages %}\n                {%- if message.role == 'system' %}\n                    {{- '<|im_start|><|system|>' + message.content + '<|im_end|>'}}\n                {%- elif message.role == 'user' %}\n                    {{- '<|im_start|><|user|>' + message.content + '<|im_end|>'}}\n                {%- elif message.role == 'assistant' %}\n                    {{- '<|im_start|><|assistant|>'}}\n                    {%- if message.content is defined %}\n                        {{- message.content}}\n                    {%- endif %}\n                    {%- if message.tool_calls is defined %}\n                        {%- for tool_call in message.tool_calls %}\n                            {%- if tool_call.function is defined %}\n                                {%- set tool_call = tool_call.function %}\n                            {%- endif %}\n                            {{- '<tool_call>' }}\n                            {{- '{' }}\n                            {{- '\"name\": \"' }}\n                            {{- tool_call.name }}\n                            {{- '\"' }}\n                            {%- if tool_call.arguments is defined %}\n                                {{- ', ' }}\n                                {{- '\"arguments\": ' }}\n                                {{- tool_call.arguments|tojson }}\n                            {%- endif %}\n                            {{- '}' }}\n                            {{- '</tool_call>' }}\n                        {%- endfor %}\n                    {%- endif %}\n                    {{- '<|im_end|>'}}\n    \n                {%- elif message.role == 'tool' %}\n                    {{- '<|im_start|><|extra_id_13|><tool_output>' + message.content + '</tool_output><|im_end|>'}}\n                {%- endif %}\n            {%- endfor %}\n            {%- if add_generation_prompt %}\n                {{- '<|im_start|><|assistant|>'}}\n            {%- endif %}",
+  "clean_up_tokenization_spaces": true,
+  "auto_map": {
+"AutoProcessor":  "processing_ax4vl.AX4VLProcessor"
+  },
+  "cls_token": "<|cls|>",
+  "eod_token": "<|endoftext|>",
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "mask_token": "<|mask|>",
+  "model_max_length": 8192,
+  "pad_token": "<|pad|>",
+  "padding_side": "right",
+  "processor_class": "AX4VLProcessor",
+  "sep_token": "<|sep|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "truncation_side": "left",
+  "unk_token": "<|unk|>",
+  "vocab_size": 102400
+}