init

.gitattributes

@@ -1,35 +1,38 @@
 *.7z filter=lfs diff=lfs merge=lfs -text
 *.arrow filter=lfs diff=lfs merge=lfs -text
 *.bin filter=lfs diff=lfs merge=lfs -text
+*.bin.* filter=lfs diff=lfs merge=lfs -text
 *.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
 *.ftz filter=lfs diff=lfs merge=lfs -text
 *.gz filter=lfs diff=lfs merge=lfs -text
 *.h5 filter=lfs diff=lfs merge=lfs -text
 *.joblib filter=lfs diff=lfs merge=lfs -text
 *.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
 *.model filter=lfs diff=lfs merge=lfs -text
 *.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
 *.onnx filter=lfs diff=lfs merge=lfs -text
 *.ot filter=lfs diff=lfs merge=lfs -text
 *.parquet filter=lfs diff=lfs merge=lfs -text
 *.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
 *.pt filter=lfs diff=lfs merge=lfs -text
 *.pth filter=lfs diff=lfs merge=lfs -text
 *.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
 saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
 *.tflite filter=lfs diff=lfs merge=lfs -text
 *.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
 *.xz filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.zstandard filter=lfs diff=lfs merge=lfs -text
+*.tfevents* filter=lfs diff=lfs merge=lfs -text
+*.db* filter=lfs diff=lfs merge=lfs -text
+*.ark* filter=lfs diff=lfs merge=lfs -text
+**/*ckpt*data* filter=lfs diff=lfs merge=lfs -text
+**/*ckpt*.meta filter=lfs diff=lfs merge=lfs -text
+**/*ckpt*.index filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.gguf* filter=lfs diff=lfs merge=lfs -text
+*.ggml filter=lfs diff=lfs merge=lfs -text
+*.llamafile* filter=lfs diff=lfs merge=lfs -text
+*.pt2 filter=lfs diff=lfs merge=lfs -text

LICENSE

@@ -0,0 +1,84 @@
+The GLM-Edge License
+
+1. 定义
+
+“许可方”是指分发其软件的 GLM-Edge 模型团队。
+“软件”是指根据本许可提供的 GLM-Edge 模型参数。
+
+2. 许可授予
+
+根据本许可的条款和条件，许可方特此授予您非排他性、全球性、不可转让、不可再许可、可撤销、免版税的版权许可。
+本许可允许您免费使用本仓库中的所有开源模型进行学术研究，对于希望将模型用于商业目的的用户，需在[这里](https://open.bigmodel.cn/mla/form)完成登记。经过登记的用户可以免费使用本模型进行商业活动，但必须遵守本许可的所有条款和条件。
+上述版权声明和本许可声明应包含在本软件的所有副本或重要部分中。
+如果您分发或提供 THUDM / 智谱AI 关于 GLM-Edge 开源模型的材料（或其任何衍生作品），或使用其中任何材料（包括 GLM-Edge 系列的所有开源模型）的产品或服务，您应:
+
+(A) 随任何此类 THUDM / 智谱AI 材料提供本协议的副本；
+(B) 在相关网站、用户界面、博客文章、关于页面或产品文档上突出显示 “Built with GLM-Edge”。
+如果您使用 THUDM / 智谱AI的 GLM-Edge 开源模型的材料来创建、训练、微调或以其他方式改进已分发或可用的 AI 模型，您还应在任何此类 AI 模型名称的开头添加 “GLM-Edge”。
+
+3. 限制
+
+您不得出于任何军事或非法目的使用、复制、修改、合并、发布、分发、复制或创建本软件的全部或部分衍生作品。
+您不得利用本软件从事任何危害国家安全和国家统一，危害社会公共利益及公序良俗，侵犯他人商业秘密、知识产权、名誉权、肖像权、财产权等权益的行为。
+您在使用中应遵循使用地所适用的法律法规政策、道德规范等要求。
+
+4. 免责声明
+
+本软件“按原样”提供，不提供任何明示或暗示的保证，包括但不限于对适销性、特定用途的适用性和非侵权性的保证。
+在任何情况下，作者或版权持有人均不对任何索赔、损害或其他责任负责，无论是在合同诉讼、侵权行为还是其他方面，由软件或软件的使用或其他交易引起、由软件引起或与之相关
+软件。
+
+5. 责任限制
+
+除适用法律禁止的范围外，在任何情况下且根据任何法律理论，无论是基于侵权行为、疏忽、合同、责任或其他原因，任何许可方均不对您承担任何直接、间接、特殊、偶然、示范性、
+或间接损害，或任何其他商业损失，即使许可人已被告知此类损害的可能性。
+
+6. 争议解决
+
+本许可受中华人民共和国法律管辖并按其解释。 因本许可引起的或与本许可有关的任何争议应提交北京市海淀区人民法院。
+请注意，许可证可能会更新到更全面的版本。 有关许可和版权的任何问题，请通过 license@zhipuai.cn 或 opensource@zhipuai.cn 与我们联系。
+
+1. Definitions
+
+“Licensor” means the GLM-Edge Model Team that distributes its Software.
+“Software” means the GLM-Edge model parameters made available under this license.
+
+2. License
+
+Under the terms and conditions of this license, the Licensor hereby grants you a non-exclusive, worldwide, non-transferable, non-sublicensable, revocable, royalty-free copyright license.
+This license allows you to use all open source models in this repository for free for academic research. For users who wish to use the models for commercial purposes, please do so [here](https://open.bigmodel.cn/mla/form)
+Complete registration. Registered users are free to use this model for commercial activities, but must comply with all terms and conditions of this license.
+The copyright notice and this license notice shall be included in all copies or substantial portions of the Software.
+If you distribute or provide THUDM / Zhipu AI materials on the GLM-Edge open source model (or any derivative works thereof), or products or services that use any materials therein (including all open source models of the GLM-Edge series), you should:
+
+(A) Provide a copy of this Agreement with any such THUDM/Zhipu AI Materials;
+(B) Prominently display "Built with GLM-Edge" on the relevant website, user interface, blog post, related page or product documentation.
+If you use materials from THUDM/ZHIPU's GLM-Edge model to create, train, operate, or otherwise improve assigned or available AI models, you should also add "GLM-Edge" to the beginning of any such AI model name.
+
+3. Restrictions
+
+You are not allowed to use, copy, modify, merge, publish, distribute, copy or create all or part of the derivative works of this software for any military or illegal purposes.
+You are not allowed to use this software to engage in any behavior that endangers national security and unity, endangers social public interests and public order, infringes on the rights and interests of others such as trade secrets, intellectual property rights, reputation rights, portrait rights, and property rights.
+You should comply with the applicable laws, regulations, policies, ethical standards, and other requirements in the place of use during use.
+
+4. Disclaimer
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE
+WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR
+COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+5. Limitation of Liability
+
+EXCEPT TO THE EXTENT PROHIBITED BY APPLICABLE LAW, IN NO EVENT AND UNDER NO LEGAL THEORY, WHETHER BASED IN TORT,
+NEGLIGENCE, CONTRACT, LIABILITY, OR OTHERWISE WILL ANY LICENSOR BE LIABLE TO YOU FOR ANY DIRECT, INDIRECT, SPECIAL,
+INCIDENTAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES, OR ANY OTHER COMMERCIAL LOSSES, EVEN IF THE LICENSOR HAS BEEN ADVISED
+OF THE POSSIBILITY OF SUCH DAMAGES.
+
+6. Dispute Resolution
+
+This license shall be governed and construed in accordance with the laws of People’s Republic of China. Any dispute
+arising from or in connection with this License shall be submitted to Haidian District People's Court in Beijing.
+
+Note that the license is subject to update to a more comprehensive version. For any questions related to the license and
+copyright, please contact us at license@zhipuai.cn.

README.md

@@ -0,0 +1,51 @@
+---
+frameworks:
+- Pytorch
+license: other
+tasks:
+- image-text-to-text
+language:
+- cn
+- en
+---
+
+# GLM-Edge-V-5B
+
+快速推理代码:
+
+```python
+import torch
+from PIL import Image
+from transformers import (
+    AutoTokenizer,
+    AutoImageProcessor,
+    AutoModelForCausalLM,
+)
+
+url = "img.png"
+messages = [{"role": "user", "content": [{"type": "image"}, {"type": "text", "text": "describe this image"}]}]
+image = Image.open(url)
+
+model_dir = "THUDM/glm-edge-v-5b"
+
+processor = AutoImageProcessor.from_pretrained(model_dir, trust_remote_code=True)
+tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained(
+    model_dir,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+    trust_remote_code=True,
+)
+
+inputs = tokenizer.apply_chat_template(
+    messages, add_generation_prompt=True, return_dict=True, tokenize=True, return_tensors="pt"
+).to(next(model.parameters()).device)
+
+generate_kwargs = {
+    **inputs,
+    "pixel_values": torch.tensor(processor(image).pixel_values).to(next(model.parameters()).device),
+}
+output = model.generate(**generate_kwargs, max_new_tokens=100)
+print(tokenizer.decode(output[0][len(inputs["input_ids"][0]):], skip_special_tokens=True))
+
+```

config.json

@@ -0,0 +1,49 @@
+{
+  "architectures": [
+    "GlmForCausalLM"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_glm.GlmConfig",
+    "AutoModel": "modeling_glm.GlmModel",
+    "AutoModelForCausalLM": "modeling_glm.GlmForCausalLM",
+    "AutoModelForSequenceClassification": "modeling_glm.GlmForSequenceClassification"
+  },
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "boi_token_id": 59256,
+  "eoi_token_id": 59257,
+  "eos_token_id": [
+    59246,
+    59253,
+    59255
+  ],
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 3072,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 4096,
+  "model_type": "glm",
+  "num_attention_heads": 24,
+  "num_hidden_layers": 40,
+  "num_key_value_heads": 6,
+  "pad_token_id": 59246,
+  "partial_rotary_factor": 1.0,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.47.0.dev0",
+  "use_cache": true,
+  "vision_config": {
+    "hidden_size": 1152,
+    "image_size": 672,
+    "intermediate_size": 4304,
+    "model_type": "siglip_vision_model",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 27,
+    "patch_size": 14,
+    "torch_dtype": "bfloat16"
+  },
+  "vocab_size": 59264
+}

configuration_glm.py

@@ -0,0 +1,152 @@
+# coding=utf-8
+# Copyright 2024 The GLM & ZhipuAI team and HuggingFace Inc. team. All rights reserved.
+#
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from transformers.configuration_utils import PretrainedConfig
+
+
+class GlmConfig(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`GlmModel`]. It is used to instantiate an Glm
+    model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
+    defaults will yield a similar configuration to that of the Glm-4-9b-chat.
+    e.g. [THUDM/glm-4-9b-chat](https://huggingface.co/THUDM/glm-4-9b-chat)
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+    Args:
+        vocab_size (`int`, *optional*, defaults to 151552):
+            Vocabulary size of the Glm model. Defines the number of different tokens that can be represented by the
+            `inputs_ids` passed when calling [`GlmModel`]
+        hidden_size (`int`, *optional*, defaults to 4096):
+            Dimension of the hidden representations.
+        intermediate_size (`int`, *optional*, defaults to 13696):
+            Dimension of the MLP representations.
+        num_hidden_layers (`int`, *optional*, defaults to 40):
+            Number of hidden layers in the Transformer decoder.
+        num_attention_heads (`int`, *optional*, defaults to 32):
+            Number of attention heads for each attention layer in the Transformer decoder.
+        num_key_value_heads (`int`, *optional*, defaults to 2):
+            This is the number of key_value heads that should be used to implement Grouped Query Attention. If
+            `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
+            `num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When
+            converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
+            by meanpooling all the original heads within that group. For more details checkout [this
+            paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to
+            `num_attention_heads`.
+        head_dim (`int`, *optional*, defaults to 128):
+            The attention head dimension.
+        hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
+            The legacy activation function. It is overwritten by the `hidden_activation`.
+        attention_dropout (`float`, *optional*, defaults to 0.0):
+            The dropout ratio for the attention probabilities.
+        max_position_embeddings (`int`, *optional*, defaults to 131072):
+            The maximum sequence length that this model might ever be used with.
+        initializer_range (`float`, *optional*, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+        rms_norm_eps (`float`, *optional*, defaults to 1.5625e-07):
+            The epsilon used by the rms normalization layers.
+        use_cache (`bool`, *optional*, defaults to `True`):
+            Whether or not the model should return the last key/values attentions (not used by all models). Only
+            relevant if `config.is_decoder=True`.
+        tie_word_embeddings (`bool`, *optional*, defaults to `False`):
+            Whether to tie weight embeddings
+        rope_theta (`float`, *optional*, defaults to 10000.0):
+            The base period of the RoPE embeddings.
+        pad_token_id (`int`, *optional*, defaults to 151329):
+            Padding token id.
+        eos_token_id (`int` | `list`, *optional*, defaults to `[151329, 151336, 151338]`):
+            End of stream token id.
+        bos_token_id (`int`, *optional*):
+            Beginning of stream token id.
+        attention_bias (`bool`, defaults to `False`, *optional*, defaults to `True`):
+            Whether to use a bias in the query, key, value and output projection layers during self-attention.
+        boi_token_id (`int`, *optional*, defaults to 151339):
+            Beginning of image token id.
+        eoi_token_id (`int` | `list`, *optional*, defaults to `[151339, 151346, 151348]`):
+            End of image token id.
+        partial_rotary_factor (`float`, *optional*, defaults to 0.5):
+            The partial rotary factor.
+        vision_config (`VisionConfig`, *optional*, defaults to `None`):
+            The vision configuration object.
+    ```python
+    >>> from transformers import GlmModel, GlmConfig
+    >>> # Initializing a Glm glm-4-9b-chat style configuration
+    >>> configuration = GlmConfig()
+    >>> # Initializing a model from the glm-4-9b-chat style configuration
+    >>> model = GlmModel(configuration)
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+    ```"""
+
+    model_type = "glm"
+    keys_to_ignore_at_inference = ["past_key_values"]
+
+    def __init__(
+        self,
+        vocab_size=65024,
+        hidden_size=4096,
+        intermediate_size=13696,
+        num_hidden_layers=28,
+        head_dim=128,
+        num_attention_heads=32,
+        max_position_embeddings=2048,
+        attention_dropout=0.0,
+        rms_norm_eps=1e-5,
+        attention_bias=False,
+        num_key_value_heads=1,
+        rope_theta=10000.0,
+        hidden_act="silu",
+        initializer_range=0.02,
+        use_cache=True,
+        tie_word_embeddings=False,
+        pad_token_id=59246,
+        bos_token_id=None,
+        eos_token_id=[59246, 59253, 59255],
+        boi_token_id=59256,
+        eoi_token_id=59257,
+        vision_config=None,
+        partial_rotary_factor=0.5,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.max_position_embeddings = max_position_embeddings
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.head_dim = head_dim
+        self.num_key_value_heads = num_key_value_heads
+        self.hidden_act = hidden_act
+        self.initializer_range = initializer_range
+        self.rms_norm_eps = rms_norm_eps
+        self.use_cache = use_cache
+        self.rope_theta = rope_theta
+        self.attention_bias = attention_bias
+        self.attention_dropout = attention_dropout
+        self.partial_rotary_factor = partial_rotary_factor
+        self.boi_token_id = boi_token_id
+        self.eoi_token_id = eoi_token_id
+        self.vision_config = vision_config
+
+        super().__init__(
+            pad_token_id=pad_token_id,
+            bos_token_id=bos_token_id,
+            eos_token_id=eos_token_id,
+            tie_word_embeddings=tie_word_embeddings,
+            **kwargs,
+        )
+
+
+__all__ = ["GlmConfig"]

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": [
+    59246,
+    59253,
+    59255
+  ],
+  "pad_token_id": 59246,
+  "transformers_version": "4.47.0.dev0"
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95d26956a9c621f4e6dd7b86452a2802176744115331f2fc7e800dbf4783b60d
+size 4969768464

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67737daa6c4aa782bc4cc4b7a7376154f48f09cc1a8267d6c13f769c3fadac42
+size 4744596440

model.safetensors.index.json

@@ -0,0 +1,788 @@
+{
+  "metadata": {
+    "total_size": 9714259840
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors",
+    "model.vision.adapter.boi": "model-00002-of-00002.safetensors",
+    "model.vision.adapter.conv.bias": "model-00002-of-00002.safetensors",
+    "model.vision.adapter.conv.weight": "model-00002-of-00002.safetensors",
+    "model.vision.adapter.eoi": "model-00002-of-00002.safetensors",
+    "model.vision.adapter.linear_proj.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
+    "model.vision.adapter.linear_proj.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
+    "model.vision.adapter.linear_proj.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.adapter.linear_proj.linear_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.adapter.linear_proj.norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.adapter.linear_proj.norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.embeddings.patch_embedding.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.embeddings.patch_embedding.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.embeddings.position_embedding.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.0.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.1.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.10.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.11.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.12.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.13.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.14.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.15.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.16.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.17.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.18.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.19.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.2.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.20.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.3.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.4.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.5.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.6.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.7.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.8.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.layer_norm1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.layer_norm1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.layer_norm2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.layer_norm2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.encoder.layers.9.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.head.attention.in_proj_bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.head.attention.in_proj_weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.head.attention.out_proj.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.head.attention.out_proj.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.head.layernorm.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.head.layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.head.mlp.fc1.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.head.mlp.fc1.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.head.mlp.fc2.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.head.mlp.fc2.weight": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.head.probe": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.post_layernorm.bias": "model-00002-of-00002.safetensors",
+    "model.vision.vit.model.vision_model.post_layernorm.weight": "model-00002-of-00002.safetensors"
+  }
+}

modeling_glm.py

@@ -0,0 +1,1327 @@
+import math
+from typing import List, Optional, Tuple, Union, Dict, Any
+
+import torch
+import torch.nn as nn
+
+from transformers.activations import ACT2FN
+from transformers.cache_utils import Cache, DynamicCache, StaticCache
+from transformers.generation import GenerationMixin
+from transformers.generation.utils import ModelOutput
+from transformers.modeling_attn_mask_utils import AttentionMaskConverter
+from transformers.modeling_flash_attention_utils import _flash_attention_forward
+from transformers.modeling_outputs import (
+    BaseModelOutputWithPast,
+    CausalLMOutputWithPast,
+    SequenceClassifierOutputWithPast,
+)
+from transformers.modeling_utils import PreTrainedModel
+from transformers.utils import (
+    add_start_docstrings,
+    add_start_docstrings_to_model_forward,
+    is_flash_attn_greater_or_equal_2_10,
+    logging,
+    replace_return_docstrings,
+)
+from transformers import __version__ as transformers_version
+
+from .siglip import VisionModel
+from .configuration_glm import GlmConfig
+
+
+logger = logging.get_logger(__name__)
+
+_CHECKPOINT_FOR_DOC = "THUDM/glm-4-9b"
+_CONFIG_FOR_DOC = "GlmConfig"
+
+
+class GlmRMSNorm(nn.Module):
+    def __init__(self, hidden_size, eps=1e-6):
+        """
+        GlmRMSNorm is equivalent to T5LayerNorm
+        """
+        super().__init__()
+        self.weight = nn.Parameter(torch.ones(hidden_size))
+        self.variance_epsilon = eps
+
+    def forward(self, hidden_states):
+        input_dtype = hidden_states.dtype
+        hidden_states = hidden_states.to(torch.float32)
+        variance = hidden_states.pow(2).mean(-1, keepdim=True)
+        hidden_states = hidden_states * torch.rsqrt(variance + self.variance_epsilon)
+        return self.weight * hidden_states.to(input_dtype)
+
+    def extra_repr(self):
+        return f"{tuple(self.weight.shape)}, eps={self.variance_epsilon}"
+
+
+class GlmRotaryEmbedding(nn.Module):
+    def __init__(self, dim, max_position_embeddings=2048, base=10000, device=None):
+        super().__init__()
+
+        self.dim = dim
+        self.max_position_embeddings = max_position_embeddings
+        self.base = base
+
+        inv_freq = 1.0 / (self.base ** (torch.arange(0, self.dim, 2, dtype=torch.int64).float() / self.dim))
+        self.register_buffer("inv_freq", tensor=inv_freq, persistent=False)
+
+    @torch.no_grad()
+    def forward(self, x, position_ids, seq_len=None):
+        # x: [bs, num_attention_heads, seq_len, head_size]
+        self.inv_freq.to(x.device)
+        inv_freq_expanded = self.inv_freq[None, :, None].float().expand(position_ids.shape[0], -1, 1)
+        position_ids_expanded = position_ids[:, None, :].float()
+        # Force float32 since bfloat16 loses precision on long contexts
+        # See https://github.com/huggingface/transformers/pull/29285
+        device_type = x.device.type
+        device_type = device_type if isinstance(device_type, str) and device_type != "mps" else "cpu"
+        with torch.autocast(device_type=device_type, enabled=False):
+            freqs = (inv_freq_expanded.float() @ position_ids_expanded.float()).transpose(1, 2)
+            emb = torch.cat((freqs, freqs), dim=-1)
+            cos = emb.cos()
+            sin = emb.sin()
+        return cos.to(dtype=x.dtype), sin.to(dtype=x.dtype)
+
+
+class GlmMLP(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+
+        self.config = config
+        self.gate_up_proj = nn.Linear(config.hidden_size, 2 * config.intermediate_size, bias=False)
+        self.down_proj = nn.Linear(config.intermediate_size, config.hidden_size, bias=False)
+
+        self.activation_fn = ACT2FN[config.hidden_act]
+
+    def forward(self, hidden_states: torch.FloatTensor) -> torch.FloatTensor:
+        up_states = self.gate_up_proj(hidden_states)
+
+        gate, up_states = up_states.chunk(2, dim=-1)
+        up_states = up_states * self.activation_fn(gate)
+
+        return self.down_proj(up_states)
+
+
+def repeat_kv(hidden_states: torch.Tensor, n_rep: int) -> torch.Tensor:
+    """
+    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
+    num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
+    """
+    batch, num_key_value_heads, slen, head_dim = hidden_states.shape
+    if n_rep == 1:
+        return hidden_states
+    hidden_states = hidden_states[:, :, None, :, :].expand(batch, num_key_value_heads, n_rep, slen, head_dim)
+    return hidden_states.reshape(batch, num_key_value_heads * n_rep, slen, head_dim)
+
+
+def rotate_half(x):
+    """Rotates half the hidden dims of the input."""
+    x1 = x[..., 0::2]
+    x2 = x[..., 1::2]
+    return torch.stack((-x2, x1), dim=-1).flatten(-2)
+
+
+def apply_rotary_pos_emb(q, k, cos, sin, position_ids=None, unsqueeze_dim=1, partial_rotary_factor=0.5):
+    """Applies Rotary Position Embedding to the query and key tensors.
+
+    Args:
+        q (`torch.Tensor`): The query tensor.
+        k (`torch.Tensor`): The key tensor.
+        cos (`torch.Tensor`): The cosine part of the rotary embedding.
+        sin (`torch.Tensor`): The sine part of the rotary embedding.
+        position_ids (`torch.Tensor`, *optional*):
+            Deprecated and unused.
+        unsqueeze_dim (`int`, *optional*, defaults to 1):
+            The 'unsqueeze_dim' argument specifies the dimension along which to unsqueeze cos[position_ids] and
+            sin[position_ids] so that they can be properly broadcasted to the dimensions of q and k. For example, note
+            that cos[position_ids] and sin[position_ids] have the shape [batch_size, seq_len, head_dim]. Then, if q and
+            k have the shape [batch_size, heads, seq_len, head_dim], then setting unsqueeze_dim=1 makes
+            cos[position_ids] and sin[position_ids] broadcastable to the shapes of q and k. Similarly, if q and k have
+            the shape [batch_size, seq_len, heads, head_dim], then set unsqueeze_dim=2.
+        partial_rotary_factor (`float`, *optional*, defaults to 0.5): The factor by which the rotary embedding.
+    Returns:
+        `tuple(torch.Tensor)` comprising of the query and key tensors rotated using the Rotary Position Embedding.
+    """
+    cos = cos.unsqueeze(unsqueeze_dim)
+    sin = sin.unsqueeze(unsqueeze_dim)
+
+    # Interleave them instead of usual shape
+    cos = cos[..., : cos.shape[-1] // 2].repeat_interleave(2, dim=-1)
+    sin = sin[..., : sin.shape[-1] // 2].repeat_interleave(2, dim=-1)
+
+    rotary_dim = int(q.shape[-1] * partial_rotary_factor)
+    q, q_pass = q[..., :rotary_dim], q[..., rotary_dim:]
+    k, k_pass = k[..., :rotary_dim], k[..., rotary_dim:]
+
+    # Apply rotary embeddings to the rotary portion
+    q = (q * cos[..., :rotary_dim]) + (rotate_half(q) * sin[..., :rotary_dim])
+    k = (k * cos[..., :rotary_dim]) + (rotate_half(k) * sin[..., :rotary_dim])
+
+    # Concatenate back the rotary and non-rotary portions
+    q_embed = torch.cat([q, q_pass], dim=-1)
+    k_embed = torch.cat([k, k_pass], dim=-1)
+
+    return q_embed, k_embed
+
+
+class GlmAttention(nn.Module):
+    """Multi-headed attention from 'Attention Is All You Need' paper"""
+
+    def __init__(self, config: GlmConfig, layer_idx: Optional[int] = None):
+        super().__init__()
+        self.config = config
+        self.layer_idx = layer_idx
+        if layer_idx is None:
+            logger.warning_once(
+                f"Instantiating {self.__class__.__name__} without passing a `layer_idx` is not recommended and will "
+                "lead to errors during the forward call if caching is used. Please make sure to provide a `layer_idx` "
+                "when creating this class."
+            )
+
+        self.attention_dropout = config.attention_dropout
+        self.hidden_size = config.hidden_size
+        self.num_heads = config.num_attention_heads
+        self.head_dim = self.hidden_size // self.num_heads
+        self.num_key_value_heads = config.num_key_value_heads
+        self.num_key_value_groups = self.num_heads // self.num_key_value_heads
+        self.is_causal = True
+        self.scaling = 1 / math.sqrt(self.head_dim)
+        self.partial_rotary_factor = config.partial_rotary_factor
+
+        if (self.head_dim * self.num_heads) != self.hidden_size:
+            raise ValueError(
+                f"hidden_size must be divisible by num_heads (got `hidden_size`: {self.hidden_size}"
+                f" and `num_heads`: {self.num_heads})."
+            )
+
+        self.q_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=config.attention_bias)
+        self.k_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=config.attention_bias)
+        self.v_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=config.attention_bias)
+        self.o_proj = nn.Linear(self.hidden_size, self.hidden_size, bias=False)
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Cache] = None,
+        output_attentions: bool = False,
+        use_cache: bool = False,
+        cache_position: Optional[torch.LongTensor] = None,
+        position_embeddings: Optional[Tuple[torch.Tensor, torch.Tensor]] = None,  # will become mandatory in v4.45
+        **kwargs,
+    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
+        bsz, q_len, _ = hidden_states.size()
+
+        query_states = self.q_proj(hidden_states)
+        key_states = self.k_proj(hidden_states)
+        value_states = self.v_proj(hidden_states)
+
+        query_states = query_states.view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
+        key_states = key_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+        value_states = value_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+
+        cos, sin = position_embeddings
+
+        query_states, key_states = apply_rotary_pos_emb(
+            query_states, key_states, cos, sin, partial_rotary_factor=self.partial_rotary_factor
+        )
+        if past_key_value is not None:
+            # sin and cos are specific to RoPE models; cache_position needed for the static cache
+            cache_kwargs = {"sin": sin, "cos": cos, "cache_position": cache_position}
+            key_states, value_states = past_key_value.update(key_states, value_states, self.layer_idx, cache_kwargs)
+
+        key_states = repeat_kv(key_states, self.num_key_value_groups)
+        value_states = repeat_kv(value_states, self.num_key_value_groups)
+
+        attn_weights = torch.matmul(query_states, key_states.transpose(2, 3)) * self.scaling
+
+        if attention_mask is not None:  # no matter the length, we just slice it
+            causal_mask = attention_mask[:, :, :, : key_states.shape[-2]]
+            attn_weights = attn_weights + causal_mask
+
+        # upcast attention to fp32
+        attn_weights = nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(query_states.dtype)
+        attn_weights = nn.functional.dropout(attn_weights, p=self.attention_dropout, training=self.training)
+        attn_output = torch.matmul(attn_weights, value_states)
+
+        if attn_output.size() != (bsz, self.num_heads, q_len, self.head_dim):
+            raise ValueError(
+                f"`attn_output` should be of size {(bsz, self.num_heads, q_len, self.head_dim)}, but is"
+                f" {attn_output.size()}"
+            )
+
+        attn_output = attn_output.transpose(1, 2).contiguous()
+
+        attn_output = attn_output.view(bsz, q_len, -1)
+        attn_output = self.o_proj(attn_output)
+
+        if not output_attentions:
+            attn_weights = None
+
+        return attn_output, attn_weights, past_key_value
+
+
+class GlmFlashAttention2(GlmAttention):
+    """
+    Glm flash attention module. This module inherits from `GlmAttention` as the weights of the module stays
+    untouched. The only required change would be on the forward pass where it needs to correctly call the public API of
+    flash attention and deal with padding tokens in case the input contains any of them.
+    """
+
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+
+        # TODO: Should be removed once Flash Attention for RoCm is bumped to 2.1.
+        # flash_attn<2.1 generates top-left aligned causal mask, while what is needed here is bottom-right alignement, that was made default for flash_attn>=2.1. This attribute is used to handle this difference. Reference: https://github.com/Dao-AILab/flash-attention/releases/tag/v2.1.0.
+        # Beware that with flash_attn<2.1, using q_seqlen != k_seqlen (except for the case q_seqlen == 1) produces a wrong mask (top-left).
+        self._flash_attn_uses_top_left_mask = not is_flash_attn_greater_or_equal_2_10()
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.LongTensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Cache] = None,
+        output_attentions: bool = False,
+        use_cache: bool = False,
+        cache_position: Optional[torch.LongTensor] = None,
+        position_embeddings: Optional[Tuple[torch.Tensor, torch.Tensor]] = None,  # will become mandatory in v4.45
+    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
+        output_attentions = False
+
+        bsz, q_len, _ = hidden_states.size()
+
+        query_states = self.q_proj(hidden_states)
+        key_states = self.k_proj(hidden_states)
+        value_states = self.v_proj(hidden_states)
+
+        # Flash attention requires the input to have the shape
+        # batch_size x seq_length x head_dim x hidden_dim
+        # therefore we just need to keep the original shape
+        query_states = query_states.view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
+        key_states = key_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+        value_states = value_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+
+        cos, sin = position_embeddings
+        query_states, key_states = apply_rotary_pos_emb(
+            query_states, key_states, cos, sin, partial_rotary_factor=self.partial_rotary_factor
+        )
+
+        if past_key_value is not None:
+            # sin and cos are specific to RoPE models; cache_position needed for the static cache
+            cache_kwargs = {"sin": sin, "cos": cos, "cache_position": cache_position}
+            key_states, value_states = past_key_value.update(key_states, value_states, self.layer_idx, cache_kwargs)
+
+        # TODO: These transpose are quite inefficient but Flash Attention requires the layout [batch_size, sequence_length, num_heads, head_dim]. We would need to refactor the KV cache
+        # to be able to avoid many of these transpose/reshape/view.
+        query_states = query_states.transpose(1, 2)
+        key_states = key_states.transpose(1, 2)
+        value_states = value_states.transpose(1, 2)
+
+        dropout_rate = self.attention_dropout if self.training else 0.0
+
+        # In PEFT, usually we cast the layer norms in float32 for training stability reasons
+        # therefore the input hidden states gets silently casted in float32. Hence, we need
+        # cast them back in the correct dtype just to be sure everything works as expected.
+        # This might slowdown training & inference so it is recommended to not cast the LayerNorms
+        # in fp32. (GlmRMSNorm handles it correctly)
+
+        input_dtype = query_states.dtype
+        if input_dtype == torch.float32:
+            if torch.is_autocast_enabled():
+                target_dtype = torch.get_autocast_gpu_dtype()
+            # Handle the case where the model is quantized
+            elif hasattr(self.config, "_pre_quantization_dtype"):
+                target_dtype = self.config._pre_quantization_dtype
+            else:
+                target_dtype = self.q_proj.weight.dtype
+
+            logger.warning_once(
+                f"The input hidden states seems to be silently casted in float32, this might be related to"
+                f" the fact you have upcasted embedding or layer norm layers in float32. We will cast back the input in"
+                f" {target_dtype}."
+            )
+
+            query_states = query_states.to(target_dtype)
+            key_states = key_states.to(target_dtype)
+            value_states = value_states.to(target_dtype)
+
+        if attention_mask is not None and len(attention_mask.shape) == 4:
+            if attention_mask.shape[1] == attention_mask.shape[2] == 1:
+                attention_mask = attention_mask.reshape(attention_mask.shape[0], -1)
+            else:
+                raise ValueError(
+                    "Get seqlens from a non-causal based full 4D attn mask is not expected. Maybe need to pass in `force_flash_attention` in `get_masks`."
+                )  # TODO
+
+        attn_output = _flash_attention_forward(
+            query_states,
+            key_states,
+            value_states,
+            attention_mask,
+            q_len,
+            position_ids=position_ids,
+            dropout=dropout_rate,
+            softmax_scale=self.scaling,
+            sliding_window=getattr(self, "sliding_window", None),
+            use_top_left_mask=self._flash_attn_uses_top_left_mask,
+            is_causal=self.is_causal,
+        )
+
+        attn_output = attn_output.reshape(bsz, q_len, -1).contiguous()
+        attn_output = self.o_proj(attn_output)
+
+        if not output_attentions:
+            attn_weights = None
+
+        return attn_output, attn_weights, past_key_value
+
+
+class GlmSdpaAttention(GlmAttention):
+    """
+    Glm attention module using torch.nn.functional.scaled_dot_product_attention. This module inherits from
+    `GlmAttention` as the weights of the module stays untouched. The only changes are on the forward pass to adapt to
+    SDPA API.
+    """
+
+    # Adapted from GlmAttention.forward
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Cache] = None,
+        output_attentions: bool = False,
+        use_cache: bool = False,
+        cache_position: Optional[torch.LongTensor] = None,
+        position_embeddings: Optional[Tuple[torch.Tensor, torch.Tensor]] = None,  # will become mandatory in v4.45
+        **kwargs,
+    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
+        if output_attentions:
+            # TODO: Improve this warning with e.g. `model.config.attn_implementation = "manual"` once this is implemented.
+            logger.warning_once(
+                "GlmModel is using GlmSdpaAttention, but `torch.nn.functional.scaled_dot_product_attention` does not support `output_attentions=True`. Falling back to the manual attention implementation, "
+                'but specifying the manual implementation will be required from Transformers version v5.0.0 onwards. This warning can be removed using the argument `attn_implementation="eager"` when loading the model.'
+            )
+            return super().forward(
+                hidden_states=hidden_states,
+                attention_mask=attention_mask,
+                position_ids=position_ids,
+                past_key_value=past_key_value,
+                output_attentions=output_attentions,
+                use_cache=use_cache,
+                cache_position=cache_position,
+                position_embeddings=position_embeddings,
+            )
+
+        bsz, q_len, _ = hidden_states.size()
+
+        query_states = self.q_proj(hidden_states)
+        key_states = self.k_proj(hidden_states)
+        value_states = self.v_proj(hidden_states)
+
+        query_states = query_states.view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
+        key_states = key_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+        value_states = value_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+
+        cos, sin = position_embeddings
+        query_states, key_states = apply_rotary_pos_emb(
+            query_states, key_states, cos, sin, partial_rotary_factor=self.partial_rotary_factor
+        )
+
+        if past_key_value is not None:
+            # sin and cos are specific to RoPE models; cache_position needed for the static cache
+            cache_kwargs = {"sin": sin, "cos": cos, "cache_position": cache_position}
+            key_states, value_states = past_key_value.update(key_states, value_states, self.layer_idx, cache_kwargs)
+
+        key_states = repeat_kv(key_states, self.num_key_value_groups)
+        value_states = repeat_kv(value_states, self.num_key_value_groups)
+
+        causal_mask = attention_mask
+        if attention_mask is not None:
+            causal_mask = causal_mask[:, :, :, : key_states.shape[-2]]
+
+        # SDPA with memory-efficient backend is currently (torch==2.1.2) bugged with non-contiguous inputs with custom attn_mask,
+        # Reference: https://github.com/pytorch/pytorch/issues/112577.
+        if query_states.device.type == "cuda" and causal_mask is not None:
+            query_states = query_states.contiguous()
+            key_states = key_states.contiguous()
+            value_states = value_states.contiguous()
+
+        # We dispatch to SDPA's Flash Attention or Efficient kernels via this `is_causal` if statement instead of an inline conditional assignment
+        # in SDPA to support both torch.compile's dynamic shapes and full graph options. An inline conditional prevents dynamic shapes from compiling.
+        is_causal = True if causal_mask is None and q_len > 1 else False
+
+        attn_output = torch.nn.functional.scaled_dot_product_attention(
+            query_states,
+            key_states,
+            value_states,
+            attn_mask=causal_mask,
+            dropout_p=self.attention_dropout if self.training else 0.0,
+            is_causal=is_causal,
+            scale=self.scaling,
+        )
+
+        attn_output = attn_output.transpose(1, 2).contiguous()
+        attn_output = attn_output.view(bsz, q_len, -1)
+
+        attn_output = self.o_proj(attn_output)
+
+        return attn_output, None, past_key_value
+
+
+GLM_ATTENTION_CLASSES = {
+    "eager": GlmAttention,
+    "flash_attention_2": GlmFlashAttention2,
+    "sdpa": GlmSdpaAttention,
+}
+
+
+class GlmDecoderLayer(nn.Module):
+    def __init__(self, config: GlmConfig, layer_idx: Optional[int] = None):
+        super().__init__()
+        self.hidden_size = config.hidden_size
+
+        # attn_implementation will not work in config.json, the correct way to use it is to pass it as
+        # a keyword argument to the model, e.g. ..from_pretrained(..., attn_implementation="flash_attention_2")
+        self.self_attn = GLM_ATTENTION_CLASSES[config._attn_implementation](config=config, layer_idx=layer_idx)
+
+        self.mlp = GlmMLP(config)
+        self.input_layernorm = GlmRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+        self.post_attention_layernorm = GlmRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Cache] = None,
+        output_attentions: Optional[bool] = False,
+        use_cache: Optional[bool] = False,
+        cache_position: Optional[torch.LongTensor] = None,
+        position_embeddings: Optional[Tuple[torch.Tensor, torch.Tensor]] = None,  # will become mandatory in v4.46
+        **kwargs,
+    ) -> Tuple[torch.FloatTensor, Optional[Tuple[torch.FloatTensor, torch.FloatTensor]]]:
+        """
+        Args:
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)`
+            attention_mask (`torch.FloatTensor`, *optional*):
+                attention mask of size `(batch_size, sequence_length)` if flash attention is used or `(batch_size, 1,
+                query_sequence_length, key_sequence_length)` if default attention is used.
+            output_attentions (`bool`, *optional*):
+                Whether or not to return the attentions tensors of all attention layers. See `attentions` under
+                returned tensors for more detail.
+            use_cache (`bool`, *optional*):
+                If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding
+                (see `past_key_values`).
+            past_key_value (`Tuple(torch.FloatTensor)`, *optional*): cached past key and value projection states
+            cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*):
+                Indices depicting the position of the input sequence tokens in the sequence
+            position_embeddings (`Tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*):
+                Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`,
+                with `head_dim` being the embedding dimension of each attention head.
+            kwargs (`dict`, *optional*):
+                Arbitrary kwargs to be ignored, used for FSDP and other methods that injects code
+                into the model
+        """
+        residual = hidden_states
+
+        hidden_states = self.input_layernorm(hidden_states)
+
+        # Self Attention
+        hidden_states, self_attn_weights, present_key_value = self.self_attn(
+            hidden_states=hidden_states,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_value=past_key_value,
+            output_attentions=output_attentions,
+            use_cache=use_cache,
+            cache_position=cache_position,
+            position_embeddings=position_embeddings,
+            **kwargs,
+        )
+        hidden_states = residual + hidden_states
+
+        # Fully Connected
+        residual = hidden_states
+        hidden_states = self.post_attention_layernorm(hidden_states)
+        hidden_states = self.mlp(hidden_states)
+        hidden_states = residual + hidden_states
+
+        outputs = (hidden_states,)
+
+        if output_attentions:
+            outputs += (self_attn_weights,)
+
+        if use_cache:
+            outputs += (present_key_value,)
+
+        return outputs
+
+
+GLM_START_DOCSTRING = r"""
+    This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the
+    library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
+    etc.)
+
+    This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
+    Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
+    and behavior.
+
+    Parameters:
+        config ([`GlmConfig`]):
+            Model configuration class with all the parameters of the model. Initializing with a config file does not
+            load the weights associated with the model, only the configuration. Check out the
+            [`~PreTrainedModel.from_pretrained`] method to load the model weights.
+"""
+
+
+@add_start_docstrings(
+    "The bare Glm Model outputting raw hidden-states without any specific head on top.",
+    GLM_START_DOCSTRING,
+)
+class GlmPreTrainedModel(PreTrainedModel):
+    config_class = GlmConfig
+    base_model_prefix = "model"
+    supports_gradient_checkpointing = True
+    _no_split_modules = ["GlmDecoderLayer"]
+    _skip_keys_device_placement = ["past_key_values"]
+    _supports_flash_attn_2 = True
+    _supports_sdpa = True
+    _supports_cache_class = True
+    _supports_quantized_cache = True
+    _supports_static_cache = True
+
+    def _init_weights(self, module):
+        std = self.config.initializer_range
+        if isinstance(module, nn.Linear):
+            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_()
+
+
+def is_empty(images_list: Optional[List[List[torch.Tensor]]]):
+    if images_list is None or len(images_list) == 0:
+        return True
+    for image_list in images_list:
+        if image_list is not None:
+            return False
+    return True
+
+
+GLM_INPUTS_DOCSTRING = r"""
+    Args:
+        input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
+            Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
+            it.
+
+            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
+            [`PreTrainedTokenizer.__call__`] for details.
+
+            [What are input IDs?](../glossary#input-ids)
+        attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
+
+            - 1 for tokens that are **not masked**,
+            - 0 for tokens that are **masked**.
+
+            [What are attention masks?](../glossary#attention-mask)
+
+            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
+            [`PreTrainedTokenizer.__call__`] for details.
+
+            If `past_key_values` is used, optionally only the last `input_ids` have to be input (see
+            `past_key_values`).
+
+            If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_attention_mask`]
+            and modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more
+            information on the default strategy.
+
+            - 1 indicates the head is **not masked**,
+            - 0 indicates the head is **masked**.
+        position_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
+            config.n_positions - 1]`.
+
+            [What are position IDs?](../glossary#position-ids)
+        past_key_values (`Cache` or `tuple(tuple(torch.FloatTensor))`, *optional*):
+            Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention
+            blocks) that can be used to speed up sequential decoding. This typically consists in the `past_key_values`
+            returned by the model at a previous stage of decoding, when `use_cache=True` or `config.use_cache=True`.
+
+            Two formats are allowed:
+            - a [`~cache_utils.Cache`] instance, see our
+            [kv cache guide](https://huggingface.co/docs/transformers/en/kv_cache);
+            - Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of
+            shape `(batch_size, num_heads, sequence_length, embed_size_per_head)`). This is also known as the legacy
+            cache format.
+
+            The model will output the same cache format that is fed as input. If no `past_key_values` are passed, the
+            legacy cache format will be returned.
+
+            If `past_key_values` are used, the user can optionally input only the last `input_ids` (those that don't
+            have their past key value states given to this model) of shape `(batch_size, 1)` instead of all `input_ids`
+            of shape `(batch_size, sequence_length)`.
+        inputs_embeds (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
+            Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This
+            is useful if you want more control over how to convert `input_ids` indices into associated vectors than the
+            model's internal embedding lookup matrix.
+        use_cache (`bool`, *optional*):
+            If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
+            `past_key_values`).
+        output_attentions (`bool`, *optional*):
+            Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned
+            tensors for more detail.
+        output_hidden_states (`bool`, *optional*):
+            Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
+            more detail.
+        return_dict (`bool`, *optional*):
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
+        cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*):
+            Indices depicting the position of the input sequence tokens in the sequence. Contrarily to `position_ids`,
+            this tensor is not affected by padding. It is used to update the cache in the correct position and to infer
+            the complete sequence length.
+"""
+
+
+@add_start_docstrings(
+    "The bare Glm Model outputting raw hidden-states without any specific head on top.",
+    GLM_START_DOCSTRING,
+)
+class GlmModel(GlmPreTrainedModel):
+    """
+    Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`GlmDecoderLayer`]
+
+    Args:
+        config: GlmConfig
+    """
+
+    def __init__(self, config: GlmConfig):
+        super().__init__(config)
+        self.padding_idx = config.pad_token_id
+        self.vocab_size = config.vocab_size
+
+        self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, self.padding_idx)
+        self.layers = nn.ModuleList(
+            [GlmDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)]
+        )
+        self.norm = GlmRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+        self.partial_rotary_factor = config.partial_rotary_factor
+        self.rotary_emb = GlmRotaryEmbedding(
+            dim=config.head_dim * self.partial_rotary_factor,
+            max_position_embeddings=config.max_position_embeddings,
+            base=config.rope_theta,
+        )
+        self.gradient_checkpointing = False
+
+        # Vision model
+        self.vision = VisionModel(config)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.embed_tokens = value
+
+    @add_start_docstrings_to_model_forward(GLM_INPUTS_DOCSTRING)
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        images: torch.Tensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Union[Cache, List[torch.FloatTensor]]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = 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,
+        **kwargs,
+    ) -> Union[Tuple, BaseModelOutputWithPast]:
+        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
+        )
+        use_cache = use_cache if use_cache is not None else self.config.use_cache
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        if (input_ids is None) ^ (inputs_embeds is not None):
+            raise ValueError("You must specify exactly one of input_ids or inputs_embeds")
+
+        if not past_key_values:
+            # not allow for inputs_embeds, because we want to process image feature
+            assert input_ids is not None and inputs_embeds is None, f"{input_ids} {inputs_embeds}"
+            inputs_embeds = self.embed_tokens(input_ids)
+            new_input_embeds = []
+            multi_flags = [True if self.config.boi_token_id in input_id.tolist() else False for input_id in input_ids]
+            images_features = None
+            if not is_empty(images):
+                images_features = self.vision(images).to(inputs_embeds.dtype)
+            image_count = 0
+            for i in range(len(input_ids)):
+                input_id = input_ids[i].tolist()
+                if multi_flags[i]:
+                    boi_token_pos = input_id.index(self.config.boi_token_id)
+                    assert boi_token_pos >= 0, "begin_of_image not found!"
+                    num_image_padding_tokens = input_id.count(self.config.boi_token_id)
+                    assert (
+                        num_image_padding_tokens == images_features[image_count].shape[0]
+                    ), f"Wrong image padding token number: {num_image_padding_tokens}"
+                    new_input_embeds.append(
+                        torch.cat(
+                            (
+                                inputs_embeds[i, :boi_token_pos],
+                                images_features[image_count].to(inputs_embeds.device),
+                                inputs_embeds[i, boi_token_pos + num_image_padding_tokens :],
+                            )
+                        )
+                    )
+                    image_count += 1
+                else:
+                    new_input_embeds.append(inputs_embeds[i])
+            inputs_embeds = torch.stack(new_input_embeds, dim=0)
+
+        if self.gradient_checkpointing and self.training and use_cache:
+            logger.warning_once(
+                "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`."
+            )
+            use_cache = False
+
+        if inputs_embeds is None:
+            if past_key_values:
+                inputs_embeds = self.embed_tokens(input_ids[:, -1:])
+            else:
+                inputs_embeds = self.embed_tokens(input_ids)
+
+        # kept for BC (non `Cache` `past_key_values` inputs)
+        return_legacy_cache = False
+        if use_cache and not isinstance(past_key_values, Cache):
+            return_legacy_cache = True
+            if past_key_values is None:
+                past_key_values = DynamicCache()
+            else:
+                past_key_values = DynamicCache.from_legacy_cache(past_key_values)
+                logger.warning_once(
+                    "We detected that you are passing `past_key_values` as a tuple of tuples. This is deprecated and "
+                    "will be removed in v4.47. Please convert your cache or use an appropriate `Cache` class "
+                    "(https://huggingface.co/docs/transformers/kv_cache#legacy-cache-format)"
+                )
+
+        if cache_position is None:
+            past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0
+            cache_position = torch.arange(
+                past_seen_tokens, past_seen_tokens + inputs_embeds.shape[1], device=inputs_embeds.device
+            )
+        if position_ids is None:
+            position_ids = cache_position.unsqueeze(0)
+
+        causal_mask = self._update_causal_mask(
+            attention_mask, inputs_embeds, cache_position, past_key_values, output_attentions
+        )
+        hidden_states = inputs_embeds
+
+        # create position embeddings to be shared across the decoder layers
+        position_embeddings = self.rotary_emb(hidden_states, position_ids)
+
+        # decoder layers
+        all_hidden_states = () if output_hidden_states else None
+        all_self_attns = () if output_attentions else None
+        next_decoder_cache = None
+
+        for decoder_layer in self.layers:
+            if output_hidden_states:
+                all_hidden_states += (hidden_states,)
+
+            if self.gradient_checkpointing and self.training:
+                layer_outputs = self._gradient_checkpointing_func(
+                    decoder_layer.__call__,
+                    hidden_states,
+                    causal_mask,
+                    position_ids,
+                    past_key_values,
+                    output_attentions,
+                    use_cache,
+                    cache_position,
+                    position_embeddings,
+                )
+            else:
+                layer_outputs = decoder_layer(
+                    hidden_states,
+                    attention_mask=causal_mask,
+                    position_ids=position_ids,
+                    past_key_value=past_key_values,
+                    output_attentions=output_attentions,
+                    use_cache=use_cache,
+                    cache_position=cache_position,
+                    position_embeddings=position_embeddings,
+                    **kwargs,
+                )
+
+            hidden_states = layer_outputs[0]
+
+            if use_cache:
+                next_decoder_cache = layer_outputs[2 if output_attentions else 1]
+
+            if output_attentions:
+                all_self_attns += (layer_outputs[1],)
+
+        hidden_states = self.norm(hidden_states)
+
+        # add hidden states from the last decoder layer
+        if output_hidden_states:
+            all_hidden_states += (hidden_states,)
+
+        next_cache = next_decoder_cache if use_cache else None
+        if return_legacy_cache:
+            next_cache = next_cache.to_legacy_cache()
+
+        if not return_dict:
+            return tuple(v for v in [hidden_states, next_cache, all_hidden_states, all_self_attns] if v is not None)
+        return BaseModelOutputWithPast(
+            last_hidden_state=hidden_states,
+            past_key_values=next_cache,
+            hidden_states=all_hidden_states,
+            attentions=all_self_attns,
+        )
+
+    def _update_causal_mask(
+        self,
+        attention_mask: torch.Tensor,
+        input_tensor: torch.Tensor,
+        cache_position: torch.Tensor,
+        past_key_values: Cache,
+        output_attentions: bool,
+    ):
+        if self.config._attn_implementation == "flash_attention_2":
+            if attention_mask is not None and 0.0 in attention_mask:
+                return attention_mask
+            return None
+
+        # For SDPA, when possible, we will rely on its `is_causal` argument instead of its `attn_mask` argument, in
+        # order to dispatch on Flash Attention 2. This feature is not compatible with static cache, as SDPA will fail
+        # to infer the attention mask.
+        past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0
+        using_static_cache = isinstance(past_key_values, StaticCache)
+
+        # When output attentions is True, sdpa implementation's forward method calls the eager implementation's forward
+        if self.config._attn_implementation == "sdpa" and not using_static_cache and not output_attentions:
+            if AttentionMaskConverter._ignore_causal_mask_sdpa(
+                attention_mask,
+                inputs_embeds=input_tensor,
+                past_key_values_length=past_seen_tokens,
+                is_training=self.training,
+            ):
+                return None
+
+        dtype, device = input_tensor.dtype, input_tensor.device
+        sequence_length = input_tensor.shape[1]
+        if using_static_cache:
+            target_length = past_key_values.get_max_cache_shape()
+        else:
+            target_length = (
+                attention_mask.shape[-1]
+                if isinstance(attention_mask, torch.Tensor)
+                else past_seen_tokens + sequence_length + 1
+            )
+
+        # In case the provided `attention` mask is 2D, we generate a causal mask here (4D).
+        causal_mask = self._prepare_4d_causal_attention_mask_with_cache_position(
+            attention_mask,
+            sequence_length=sequence_length,
+            target_length=target_length,
+            dtype=dtype,
+            device=device,
+            cache_position=cache_position,
+            batch_size=input_tensor.shape[0],
+        )
+
+        if (
+            self.config._attn_implementation == "sdpa"
+            and attention_mask is not None
+            and attention_mask.device.type == "cuda"
+            and not output_attentions
+        ):
+            # Attend to all tokens in fully masked rows in the causal_mask, for example the relevant first rows when
+            # using left padding. This is required by F.scaled_dot_product_attention memory-efficient attention path.
+            # Details: https://github.com/pytorch/pytorch/issues/110213
+            min_dtype = torch.finfo(dtype).min
+            causal_mask = AttentionMaskConverter._unmask_unattended(causal_mask, min_dtype)
+
+        return causal_mask
+
+    @staticmethod
+    def _prepare_4d_causal_attention_mask_with_cache_position(
+        attention_mask: torch.Tensor,
+        sequence_length: int,
+        target_length: int,
+        dtype: torch.dtype,
+        device: torch.device,
+        cache_position: torch.Tensor,
+        batch_size: int,
+        **kwargs,
+    ):
+        """
+        Creates a causal 4D mask of shape `(batch_size, 1, query_length, key_value_length)` from a 2D mask of shape
+        `(batch_size, key_value_length)`, or if the input `attention_mask` is already 4D, do nothing.
+
+        Args:
+            attention_mask (`torch.Tensor`):
+                A 2D attention mask of shape `(batch_size, key_value_length)` or a 4D attention mask of shape
+                `(batch_size, 1, query_length, key_value_length)`.
+            sequence_length (`int`):
+                The sequence length being processed.
+            target_length (`int`):
+                The target length: when generating with static cache, the mask should be as long as the static cache,
+                to account for the 0 padding, the part of the cache that is not filled yet.
+            dtype (`torch.dtype`):
+                The dtype to use for the 4D attention mask.
+            device (`torch.device`):
+                The device to plcae the 4D attention mask on.
+            cache_position (`torch.Tensor`):
+                Indices depicting the position of the input sequence tokens in the sequence.
+            batch_size (`torch.Tensor`):
+                Batch size.
+        """
+        if attention_mask is not None and attention_mask.dim() == 4:
+            # In this case we assume that the mask comes already in inverted form and requires no inversion or slicing.
+            causal_mask = attention_mask
+        else:
+            min_dtype = torch.finfo(dtype).min
+            causal_mask = torch.full(
+                (sequence_length, target_length), fill_value=min_dtype, dtype=dtype, device=device
+            )
+            if sequence_length != 1:
+                causal_mask = torch.triu(causal_mask, diagonal=1)
+            causal_mask *= torch.arange(target_length, device=device) > cache_position.reshape(-1, 1)
+            causal_mask = causal_mask[None, None, :, :].expand(batch_size, 1, -1, -1)
+            if attention_mask is not None:
+                causal_mask = causal_mask.clone()  # copy to contiguous memory for in-place edit
+                mask_length = attention_mask.shape[-1]
+                padding_mask = causal_mask[:, :, :, :mask_length] + attention_mask[:, None, None, :]
+                padding_mask = padding_mask == 0
+                causal_mask[:, :, :, :mask_length] = causal_mask[:, :, :, :mask_length].masked_fill(
+                    padding_mask, min_dtype
+                )
+
+        return causal_mask
+
+
+class GlmForCausalLM(GlmPreTrainedModel, GenerationMixin):
+    _tied_weights_keys = ["lm_head.weight"]
+
+    def __init__(self, config: GlmConfig):
+        super().__init__(config)
+        self.model = GlmModel(config)
+        self.vocab_size = config.vocab_size
+        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.model.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.model.embed_tokens = value
+
+    def get_output_embeddings(self):
+        return self.lm_head
+
+    def set_output_embeddings(self, new_embeddings):
+        self.lm_head = new_embeddings
+
+    def set_decoder(self, decoder):
+        self.model = decoder
+
+    def get_decoder(self):
+        return self.model
+
+    def _update_model_kwargs_for_generation(
+        self,
+        outputs: ModelOutput,
+        model_kwargs: Dict[str, Any],
+        is_encoder_decoder: bool = False,
+        standardize_cache_format: bool = False,
+    ) -> Dict[str, Any]:
+        # update past_key_values
+        if int(transformers_version.split(".")[1]) >= 44:
+            assert not standardize_cache_format
+            _, cache = self._extract_past_from_model_output(outputs)
+            model_kwargs["past_key_values"] = cache
+        else:
+            cache = self._extract_past_from_model_output(outputs, standardize_cache_format)
+
+        # update attention mask
+        if "attention_mask" in model_kwargs:
+            attention_mask = model_kwargs["attention_mask"]
+            model_kwargs["attention_mask"] = torch.cat(
+                [attention_mask, attention_mask.new_ones((attention_mask.shape[0], 1))], dim=-1
+            )
+
+        # update position ids
+        if "position_ids" in model_kwargs:
+            position_ids = model_kwargs["position_ids"]
+            new_position_id = position_ids[..., -1:].clone()
+            new_position_id += 1
+            model_kwargs["position_ids"] = torch.cat([position_ids, new_position_id], dim=-1)
+
+        model_kwargs["is_first_forward"] = False
+        return model_kwargs
+
+    def _create_position_ids_from_attention_mask(self, attention_mask):
+        # Initialize a tensor of the same shape as attention_mask to hold position IDs
+        position_ids = torch.zeros_like(attention_mask, dtype=torch.long, device=attention_mask.device)
+        # Iterate over the batch
+        for i, mask in enumerate(attention_mask):
+            # Find the positions where the mask is 1
+            positions = torch.nonzero(mask, as_tuple=False).squeeze(1).to(attention_mask.device)
+            # Assign position IDs to those positions
+            position_ids[i, positions] = torch.arange(start=0, end=positions.size(0), dtype=torch.long).to(
+                attention_mask.device
+            )
+        return position_ids
+
+    def prepare_inputs_for_generation(
+        self,
+        input_ids: torch.LongTensor,
+        pixel_values: Optional[torch.Tensor] = torch.zeros([1, 1, 1, 3, 672, 672]),
+        past_key_values: Optional[torch.Tensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.Tensor] = None,
+        use_cache: Optional[bool] = None,
+        is_first_forward: bool = True,
+        **kwargs,
+    ) -> dict:
+        if position_ids is None:
+            if attention_mask is None:
+                # Can only build sequential ids. Raise error right now
+                raise ValueError("Cannot create position ids when attention mask is None")
+            else:
+                position_ids = self._create_position_ids_from_attention_mask(attention_mask)
+        if not is_first_forward:
+            if past_key_values is not None:
+                position_ids = position_ids[..., -1:]
+                input_ids = input_ids[:, -1:]
+        return {
+            "input_ids": input_ids,
+            "pixel_values": pixel_values,
+            "past_key_values": past_key_values,
+            "position_ids": position_ids,
+            "attention_mask": attention_mask,
+            "return_last_logit": True,
+            "use_cache": use_cache,
+        }
+
+    @add_start_docstrings_to_model_forward(GLM_INPUTS_DOCSTRING)
+    @replace_return_docstrings(output_type=CausalLMOutputWithPast, config_class=_CONFIG_FOR_DOC)
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        pixel_values: torch.Tensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Union[Cache, List[torch.FloatTensor]]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = 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,
+        num_logits_to_keep: int = 0,
+        **loss_kwargs,
+    ) -> Union[Tuple, CausalLMOutputWithPast]:
+        r"""
+        Args:
+            labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+                Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
+                config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
+                (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
+
+            num_logits_to_keep (`int`, *optional*):
+                Calculate logits for the last `num_logits_to_keep` tokens. If `0`, calculate logits for all
+                `input_ids` (special case). Only last token logits are needed for generation, and calculating them only for that
+                token can save memory, which becomes pretty significant for long sequences or large vocabulary size.
+
+        Returns:
+
+        Example:
+
+        ```python
+        >>> from transformers import AutoTokenizer, GlmForCausalLM
+
+        >>> model = GlmForCausalLM.from_pretrained("THUDM/glm-4v-9b")
+        >>> tokenizer = AutoTokenizer.from_pretrained("THUDm/glm-4v-9b")
+
+        >>> prompt = "What is your favorite condiment?"
+        >>> inputs = tokenizer(prompt, return_tensors="pt")
+
+        >>> # Generate
+        >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
+        >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
+        "What is your favorite condiment?"
+        ```"""
+        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
+        batch_size, num_concurrent_media, num_tiles, num_channels, height, width = pixel_values.shape
+        pixel_values = pixel_values.reshape(batch_size * num_concurrent_media * num_tiles, num_channels, height, width)
+
+        # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
+        outputs = self.model(
+            input_ids=input_ids,
+            images=pixel_values,
+            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,
+        )
+
+        hidden_states = outputs[0]
+        # Only compute necessary logits, and do not upcast them to float if we are not computing the loss
+        logits = self.lm_head(hidden_states[:, -num_logits_to_keep:, :])
+
+        loss = None
+        if labels is not None:
+            loss = self.loss_function(logits, labels, self.vocab_size, **loss_kwargs)
+
+        if not return_dict:
+            output = (logits,) + outputs[1:]
+            return (loss,) + output if loss is not None else output
+
+        return CausalLMOutputWithPast(
+            loss=loss,
+            logits=logits,
+            past_key_values=outputs.past_key_values,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+
+@add_start_docstrings(
+    """
+    The Glm Model transformer with a sequence classification head on top (linear layer).
+
+    [`vForSequenceClassification`] uses the last token in order to do the classification, as other causal models
+    (e.g. GPT-2) do.
+
+    Since it does classification on the last token, it requires to know the position of the last token. If a
+    `pad_token_id` is defined in the configuration, it finds the last token that is not a padding token in each row. If
+    no `pad_token_id` is defined, it simply takes the last value in each row of the batch. Since it cannot guess the
+    padding tokens when `inputs_embeds` are passed instead of `input_ids`, it does the same (take the last value in
+    each row of the batch).
+    """,
+    GLM_START_DOCSTRING,
+)
+class GlmForSequenceClassification(GlmPreTrainedModel):
+    def __init__(self, config: GlmConfig):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+        self.model = GlmModel(config)
+        self.score = nn.Linear(config.hidden_size, self.num_labels, bias=False)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.model.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.model.embed_tokens = value
+
+    @add_start_docstrings_to_model_forward(GLM_INPUTS_DOCSTRING)
+    def forward(
+        self,
+        input_ids: Optional[torch.LongTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Union[Cache, List[torch.FloatTensor]]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = 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,
+    ) -> Union[Tuple, SequenceClassifierOutputWithPast]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
+            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
+            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        transformer_outputs = self.model(
+            input_ids,
+            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,
+        )
+        hidden_states = transformer_outputs[0]
+        logits = self.score(hidden_states)
+
+        if input_ids is not None:
+            batch_size = input_ids.shape[0]
+        else:
+            batch_size = inputs_embeds.shape[0]
+
+        if self.config.pad_token_id is None and batch_size != 1:
+            raise ValueError("Cannot handle batch sizes > 1 if no padding token is defined.")
+        if self.config.pad_token_id is None:
+            sequence_lengths = -1
+        else:
+            if input_ids is not None:
+                # if no pad token found, use modulo instead of reverse indexing for ONNX compatibility
+                sequence_lengths = torch.eq(input_ids, self.config.pad_token_id).int().argmax(-1) - 1
+                sequence_lengths = sequence_lengths % input_ids.shape[-1]
+                sequence_lengths = sequence_lengths.to(logits.device)
+            else:
+                sequence_lengths = -1
+
+        pooled_logits = logits[torch.arange(batch_size, device=logits.device), sequence_lengths]
+
+        loss = None
+        if labels is not None:
+            loss = self.loss_function(logits=logits, labels=labels, pooled_logits=pooled_logits, config=self.config)
+
+        if not return_dict:
+            output = (pooled_logits,) + transformer_outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return SequenceClassifierOutputWithPast(
+            loss=loss,
+            logits=pooled_logits,
+            past_key_values=transformer_outputs.past_key_values,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+        )
+
+
+__all__ = [
+    "GlmPreTrainedModel",
+    "GlmModel",
+    "GlmForCausalLM",
+    "GlmForSequenceClassification",
+]

preprocessor_config.json

@@ -0,0 +1,25 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_pad": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "MllamaImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "max_image_tiles": 1,
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 672,
+    "width": 672
+  }
+}

siglip.py

@@ -0,0 +1,67 @@
+import torch
+from torch import nn
+from transformers import SiglipVisionModel, SiglipVisionConfig
+
+# 384/14=27.428571428571427 is not an integer, so the actual pos embedding is 729, sqrt(729)*14=378. So the implementation uses the floor
+
+class SiglipEncoder(nn.Module):
+    def __init__(self, vision_config):
+        super(SiglipEncoder, self).__init__()
+
+        config = SiglipVisionConfig(**vision_config)
+        self.model = SiglipVisionModel(config)
+
+    def forward(self, images):
+        outputs = self.model(images).last_hidden_state
+        return outputs
+
+
+class GLU(nn.Module):
+    def __init__(self, args, in_features):
+        super().__init__()
+        self.linear_proj = nn.Linear(in_features, args.hidden_size, bias=False)
+        self.norm1 = nn.LayerNorm(args.hidden_size)
+        self.act1 = nn.GELU()
+        self.act2 = nn.functional.silu
+        self.dense_h_to_4h = nn.Linear(args.hidden_size, args.intermediate_size, bias=False)
+        self.gate_proj = nn.Linear(args.hidden_size, args.intermediate_size, bias=False)
+        self.dense_4h_to_h = nn.Linear(args.intermediate_size, args.hidden_size, bias=False)
+
+    def forward(self, x):
+        x = self.linear_proj(x)
+        x = self.act1(self.norm1(x))
+        x = self.act2(self.gate_proj(x)) * self.dense_h_to_4h(x)
+        x = self.dense_4h_to_h(x)
+        return x
+
+
+class Adapter(nn.Module):
+    def __init__(self, eva_hidden_size, args):
+        super().__init__()
+        self.boi = nn.Parameter(torch.ones(1, 1, args.hidden_size).float())
+        self.eoi = nn.Parameter(torch.ones(1, 1, args.hidden_size).float())
+        self.conv = nn.Conv2d(in_channels=eva_hidden_size, out_channels=args.hidden_size, kernel_size=2, stride=2)
+        self.linear_proj = GLU(args, args.hidden_size)
+
+    def forward(self, image_emb):
+        b, s, e = image_emb.shape # (b, 6400, 1792)
+        grid_size = int(s**0.5)
+        image_emb = image_emb.view(b, grid_size, grid_size, e).permute(0,3,1,2) # (b, 1792, 80, 80)
+        image_emb = self.conv(image_emb) # (b, 4096, 40, 40)
+        image_emb = image_emb.flatten(2).transpose(1, 2) # (b, 1600, 4096)
+        image_emb = self.linear_proj(image_emb) # (b, 1600, 6656)
+        image_emb = torch.cat([self.boi.repeat(len(image_emb), 1, 1), image_emb, self.eoi.repeat(len(image_emb), 1, 1)], dim=1)
+        return image_emb
+
+
+class VisionModel(torch.nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.dtype = config.torch_dtype
+        self.vit = SiglipEncoder(config.vision_config)
+        self.adapter = Adapter(config.vision_config['hidden_size'], config)
+
+    def forward(self, image):
+        image = image.to(self.dtype)
+        vit_output = self.vit(image)
+        return self.adapter(vit_output).to(self.dtype)

special_tokens_map.json

@@ -0,0 +1,32 @@
+{
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "[MASK]",
+    "[gMASK]",
+    "[sMASK]",
+    "<sop>",
+    "<eop>",
+    "<|system|>",
+    "<|user|>",
+    "<|assistant|>",
+    "<|observation|>",
+    "<|begin_of_image|>",
+    "<|end_of_image|>",
+    "<|begin_of_video|>",
+    "<|end_of_video|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,143 @@
+{
+  "added_tokens_decoder": {
+    "59246": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59247": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59248": {
+      "content": "[gMASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59249": {
+      "content": "[sMASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59250": {
+      "content": "<sop>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59251": {
+      "content": "<eop>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59252": {
+      "content": "<|system|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59253": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59254": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59255": {
+      "content": "<|observation|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59256": {
+      "content": "<|begin_of_image|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59257": {
+      "content": "<|end_of_image|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59258": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "59259": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "[MASK]",
+    "[gMASK]",
+    "[sMASK]",
+    "<sop>",
+    "<eop>",
+    "<|system|>",
+    "<|user|>",
+    "<|assistant|>",
+    "<|observation|>",
+    "<|begin_of_image|>"
+  ],
+  "chat_template": "{% for item in messages %}{% if item['role'] != 'system' %}<|{{ item['role'] }}|>\n{% for content in item['content'] %}{% if content['type'] == 'image' %}{% for _ in range(578) %}<|begin_of_image|>{% endfor %}{% elif content['type'] == 'text' %}{{ content['text'] }}{% endif %}{% endfor %}\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|assistant|>\n{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "do_lower_case": false,
+  "eos_token": "<|user|>",
+  "image_size": 672,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 8192,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "remove_space": false,
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}