Initial commit

.gitattributes

@@ -1,35 +1,2 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow 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
+.safetensors filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,83 @@
+---
+base_model: Qwen/Qwen-VL-Chat
+tags:
+- qwen-vl-chat
+- qwen
+- dhenu
+- lora
+- conversational
+- finetune
+- chatml
+- synthetic data
+license: apache-2.0
+language:
+- en
+---
+# Dhenu Vision LoRA 0.1
+
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/616d9181c3bac80637586601/sDak407sDNB94HmX4Cxo-.png)
+
+*Inspired by the mythical Kaamdhenu, the wish-fulfilling cow of Hindu mythology, we, at [KissanAI](https://kissan.ai), are introducing a series of finetuned Language Learning Models (LLM), "Dhenu", designed for the agricultural sector. Dhenu merges the depth of agricultural practices with modern AI capabilities, aimed at enriching the farming community with actionable insights. This model embodies the fusion of tradition and technology, offering tailored guidance to navigate the complexities of agriculture with precision and ease. Dhenu stands as a testament to innovation, guiding farmers towards a prosperous and sustainable future.*
+
+## Introduction
+
+Dhenu-vision-lora-v0.1 is an open-source agricultural disease detection model fine-tuned on the Qwen-VL-chat model. It is specifically designed to assist with diseases among three major crops, rice, maize, and wheat, in conversational way. We incorporate Low Rank adaptation techniques for low-cost fine-tuning on agricultural datasets.
+
+We observed that even with lora, the model is performing **2X** better over the base model for the objectives mentioned below.
+
+## Model details
+
+**Model type:**
+
+Base LLM: "Qwen/Qwen-VL-Chat"
+Method: LoRA
+
+**Model date:**
+
+Dhenu-vision-lora-v0.1 was trained in 03-2024.
+
+**Resources for more information:**
+
+[https://huggingface.co/Qwen/Qwen-VL-Chat](https://huggingface.co/Qwen/Qwen-VL-Chat)
+
+[https://github.com/QwenLM/Qwen-VL](https://github.com/QwenLM/Qwen-VL)
+
+## Dataset
+
+Syntehtic dataset of ~9000 images for three major crops, **Maize**, **Rice**, and **Wheat**, for following common disease identifiable from leaves:
+
+- Leaf blight
+- Leaf spot
+- Streak virus
+- Bacterial blight
+- Blast
+- Tungro
+- Crown and Root Rot
+- Leaf rust
+- Wheat loose smut
+
+For objectives:
+
+- Disease identification
+- Symptoms
+- Cure
+- Prevention methods
+- Severity 
+
+## Evaluation
+
+For a collection of 500 images evaluated using multiple prompt-based approaches, we achieved the following results:
+
+Qwen-VL-chat: 17.95%
+Dhenu-vision-lora-0.1: 36.13%
+GPT4-V: 51.59%
+
+## License
+
+[Qwen-VL-chat](https://huggingface.co/Qwen/Qwen-VL-Chat/blob/main/LICENSE) license
+
+## ToDo
+
+[ ] SFT model
+
+[ ] 15+ crops and 80 diseases

config.json

@@ -0,0 +1,49 @@
+{
+  "_name_or_path": "Qwen/Qwen-VL-Chat",
+  "architectures": [
+    "QWenLMHeadModel"
+  ],
+  "attn_dropout_prob": 0.0,
+  "auto_map": {
+    "AutoConfig": "Qwen/Qwen-VL-Chat--configuration_qwen.QWenConfig",
+    "AutoModelForCausalLM": "Qwen/Qwen-VL-Chat--modeling_qwen.QWenLMHeadModel"
+  },
+  "bf16": true,
+  "emb_dropout_prob": 0.0,
+  "fp16": false,
+  "fp32": false,
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 22016,
+  "kv_channels": 128,
+  "layer_norm_epsilon": 1e-06,
+  "max_position_embeddings": 8192,
+  "model_type": "qwen",
+  "no_bias": true,
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "onnx_safe": null,
+  "rotary_emb_base": 10000,
+  "rotary_pct": 1.0,
+  "scale_attn_weights": true,
+  "seq_length": 2048,
+  "tie_word_embeddings": false,
+  "tokenizer_type": "QWenTokenizer",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.32.0",
+  "use_cache": true,
+  "use_dynamic_ntk": true,
+  "use_flash_attn": false,
+  "use_logn_attn": true,
+  "visual": {
+    "heads": 16,
+    "image_size": 448,
+    "image_start_id": 151857,
+    "layers": 48,
+    "mlp_ratio": 4.9231,
+    "output_dim": 4096,
+    "patch_size": 14,
+    "width": 1664
+  },
+  "vocab_size": 151936
+}

generation_config.json

@@ -0,0 +1,11 @@
+{
+  "chat_format": "chatml",
+  "do_sample": true,
+  "eos_token_id": 151643,
+  "max_new_tokens": 512,
+  "max_window_size": 6144,
+  "pad_token_id": 151643,
+  "top_k": 0,
+  "top_p": 0.3,
+  "transformers_version": "4.32.0"
+}

model-00001-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27e05e8e11017ec1adfb7b7465c915975c463e614f9826b5d97455fc6ab4e2b8
+size 1964066488

model-00002-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9387f2dc898c11b0f05af6eb2d877c737a337b8942052d60c1a80152d1e00e7d
+size 2023960808

model-00003-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe4972e9280557c202ed63c7d1f69295580b0ee493b3e66aff25d1eeb2de4ee9
+size 2023960816

model-00004-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3380299fbc618e8df5bf86b7e85de1819956314ecd9bd622bd2b8af790dedab0
+size 2023960848

model-00005-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46308122ed427cd3a0ca330f5b6a3e90b05dd915f02809ffb0262313e00d8822
+size 2023960848

model-00006-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc152a8cccb70413189214db8e59fe28a64d5d8fd1b3f3c88836d3214eccb0bd
+size 2023960848

model-00007-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90e4de937cb6fc7046784ace01ed763e92ec30a6e780cef3cb781fe1fdb014f5
+size 2023960848

model-00008-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:707318bf40be16efd6d87a8fc58c86e42deacdb13c4616c40f50fce92b9c2046
+size 2044696480

model-00009-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54e25732179f8ebcd2843e46a31520d655011535b306541b0a164b8c06396f88
+size 1916789392

model-00010-of-00010.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d6b1314f3680e32ce7c69a02f17aed4c1c7bd174269f97b790ef871db10974e
+size 1244659840

model.safetensors.index.json

@@ -0,0 +1,860 @@
+{
+  "metadata": {
+    "total_size": 19313870336
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00010-of-00010.safetensors",
+    "transformer.h.0.attn.c_attn.bias": "model-00001-of-00010.safetensors",
+    "transformer.h.0.attn.c_attn.weight": "model-00001-of-00010.safetensors",
+    "transformer.h.0.attn.c_proj.weight": "model-00001-of-00010.safetensors",
+    "transformer.h.0.ln_1.weight": "model-00001-of-00010.safetensors",
+    "transformer.h.0.ln_2.weight": "model-00001-of-00010.safetensors",
+    "transformer.h.0.mlp.c_proj.weight": "model-00001-of-00010.safetensors",
+    "transformer.h.0.mlp.w1.weight": "model-00001-of-00010.safetensors",
+    "transformer.h.0.mlp.w2.weight": "model-00001-of-00010.safetensors",
+    "transformer.h.1.attn.c_attn.bias": "model-00001-of-00010.safetensors",
+    "transformer.h.1.attn.c_attn.weight": "model-00001-of-00010.safetensors",
+    "transformer.h.1.attn.c_proj.weight": "model-00001-of-00010.safetensors",
+    "transformer.h.1.ln_1.weight": "model-00001-of-00010.safetensors",
+    "transformer.h.1.ln_2.weight": "model-00001-of-00010.safetensors",
+    "transformer.h.1.mlp.c_proj.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.1.mlp.w1.weight": "model-00001-of-00010.safetensors",
+    "transformer.h.1.mlp.w2.weight": "model-00001-of-00010.safetensors",
+    "transformer.h.10.attn.c_attn.bias": "model-00003-of-00010.safetensors",
+    "transformer.h.10.attn.c_attn.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.10.attn.c_proj.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.10.ln_1.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.10.ln_2.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.10.mlp.c_proj.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.10.mlp.w1.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.10.mlp.w2.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.11.attn.c_attn.bias": "model-00003-of-00010.safetensors",
+    "transformer.h.11.attn.c_attn.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.11.attn.c_proj.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.11.ln_1.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.11.ln_2.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.11.mlp.c_proj.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.11.mlp.w1.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.11.mlp.w2.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.12.attn.c_attn.bias": "model-00004-of-00010.safetensors",
+    "transformer.h.12.attn.c_attn.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.12.attn.c_proj.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.12.ln_1.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.12.ln_2.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.12.mlp.c_proj.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.12.mlp.w1.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.12.mlp.w2.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.13.attn.c_attn.bias": "model-00004-of-00010.safetensors",
+    "transformer.h.13.attn.c_attn.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.13.attn.c_proj.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.13.ln_1.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.13.ln_2.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.13.mlp.c_proj.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.13.mlp.w1.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.13.mlp.w2.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.14.attn.c_attn.bias": "model-00004-of-00010.safetensors",
+    "transformer.h.14.attn.c_attn.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.14.attn.c_proj.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.14.ln_1.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.14.ln_2.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.14.mlp.c_proj.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.14.mlp.w1.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.14.mlp.w2.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.15.attn.c_attn.bias": "model-00004-of-00010.safetensors",
+    "transformer.h.15.attn.c_attn.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.15.attn.c_proj.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.15.ln_1.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.15.ln_2.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.15.mlp.c_proj.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.15.mlp.w1.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.15.mlp.w2.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.16.attn.c_attn.bias": "model-00004-of-00010.safetensors",
+    "transformer.h.16.attn.c_attn.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.16.attn.c_proj.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.16.ln_1.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.16.ln_2.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.16.mlp.c_proj.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.16.mlp.w1.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.16.mlp.w2.weight": "model-00004-of-00010.safetensors",
+    "transformer.h.17.attn.c_attn.bias": "model-00005-of-00010.safetensors",
+    "transformer.h.17.attn.c_attn.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.17.attn.c_proj.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.17.ln_1.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.17.ln_2.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.17.mlp.c_proj.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.17.mlp.w1.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.17.mlp.w2.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.18.attn.c_attn.bias": "model-00005-of-00010.safetensors",
+    "transformer.h.18.attn.c_attn.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.18.attn.c_proj.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.18.ln_1.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.18.ln_2.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.18.mlp.c_proj.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.18.mlp.w1.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.18.mlp.w2.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.19.attn.c_attn.bias": "model-00005-of-00010.safetensors",
+    "transformer.h.19.attn.c_attn.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.19.attn.c_proj.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.19.ln_1.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.19.ln_2.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.19.mlp.c_proj.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.19.mlp.w1.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.19.mlp.w2.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.2.attn.c_attn.bias": "model-00002-of-00010.safetensors",
+    "transformer.h.2.attn.c_attn.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.2.attn.c_proj.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.2.ln_1.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.2.ln_2.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.2.mlp.c_proj.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.2.mlp.w1.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.2.mlp.w2.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.20.attn.c_attn.bias": "model-00005-of-00010.safetensors",
+    "transformer.h.20.attn.c_attn.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.20.attn.c_proj.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.20.ln_1.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.20.ln_2.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.20.mlp.c_proj.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.20.mlp.w1.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.20.mlp.w2.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.21.attn.c_attn.bias": "model-00005-of-00010.safetensors",
+    "transformer.h.21.attn.c_attn.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.21.attn.c_proj.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.21.ln_1.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.21.ln_2.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.21.mlp.c_proj.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.21.mlp.w1.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.21.mlp.w2.weight": "model-00005-of-00010.safetensors",
+    "transformer.h.22.attn.c_attn.bias": "model-00006-of-00010.safetensors",
+    "transformer.h.22.attn.c_attn.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.22.attn.c_proj.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.22.ln_1.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.22.ln_2.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.22.mlp.c_proj.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.22.mlp.w1.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.22.mlp.w2.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.23.attn.c_attn.bias": "model-00006-of-00010.safetensors",
+    "transformer.h.23.attn.c_attn.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.23.attn.c_proj.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.23.ln_1.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.23.ln_2.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.23.mlp.c_proj.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.23.mlp.w1.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.23.mlp.w2.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.24.attn.c_attn.bias": "model-00006-of-00010.safetensors",
+    "transformer.h.24.attn.c_attn.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.24.attn.c_proj.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.24.ln_1.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.24.ln_2.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.24.mlp.c_proj.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.24.mlp.w1.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.24.mlp.w2.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.25.attn.c_attn.bias": "model-00006-of-00010.safetensors",
+    "transformer.h.25.attn.c_attn.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.25.attn.c_proj.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.25.ln_1.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.25.ln_2.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.25.mlp.c_proj.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.25.mlp.w1.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.25.mlp.w2.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.26.attn.c_attn.bias": "model-00006-of-00010.safetensors",
+    "transformer.h.26.attn.c_attn.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.26.attn.c_proj.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.26.ln_1.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.26.ln_2.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.26.mlp.c_proj.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.26.mlp.w1.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.26.mlp.w2.weight": "model-00006-of-00010.safetensors",
+    "transformer.h.27.attn.c_attn.bias": "model-00007-of-00010.safetensors",
+    "transformer.h.27.attn.c_attn.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.27.attn.c_proj.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.27.ln_1.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.27.ln_2.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.27.mlp.c_proj.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.27.mlp.w1.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.27.mlp.w2.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.28.attn.c_attn.bias": "model-00007-of-00010.safetensors",
+    "transformer.h.28.attn.c_attn.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.28.attn.c_proj.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.28.ln_1.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.28.ln_2.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.28.mlp.c_proj.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.28.mlp.w1.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.28.mlp.w2.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.29.attn.c_attn.bias": "model-00007-of-00010.safetensors",
+    "transformer.h.29.attn.c_attn.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.29.attn.c_proj.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.29.ln_1.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.29.ln_2.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.29.mlp.c_proj.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.29.mlp.w1.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.29.mlp.w2.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.3.attn.c_attn.bias": "model-00002-of-00010.safetensors",
+    "transformer.h.3.attn.c_attn.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.3.attn.c_proj.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.3.ln_1.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.3.ln_2.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.3.mlp.c_proj.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.3.mlp.w1.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.3.mlp.w2.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.30.attn.c_attn.bias": "model-00007-of-00010.safetensors",
+    "transformer.h.30.attn.c_attn.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.30.attn.c_proj.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.30.ln_1.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.30.ln_2.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.30.mlp.c_proj.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.30.mlp.w1.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.30.mlp.w2.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.31.attn.c_attn.bias": "model-00007-of-00010.safetensors",
+    "transformer.h.31.attn.c_attn.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.31.attn.c_proj.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.31.ln_1.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.31.ln_2.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.31.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.h.31.mlp.w1.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.31.mlp.w2.weight": "model-00007-of-00010.safetensors",
+    "transformer.h.4.attn.c_attn.bias": "model-00002-of-00010.safetensors",
+    "transformer.h.4.attn.c_attn.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.4.attn.c_proj.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.4.ln_1.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.4.ln_2.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.4.mlp.c_proj.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.4.mlp.w1.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.4.mlp.w2.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.5.attn.c_attn.bias": "model-00002-of-00010.safetensors",
+    "transformer.h.5.attn.c_attn.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.5.attn.c_proj.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.5.ln_1.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.5.ln_2.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.5.mlp.c_proj.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.5.mlp.w1.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.5.mlp.w2.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.6.attn.c_attn.bias": "model-00002-of-00010.safetensors",
+    "transformer.h.6.attn.c_attn.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.6.attn.c_proj.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.6.ln_1.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.6.ln_2.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.6.mlp.c_proj.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.6.mlp.w1.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.6.mlp.w2.weight": "model-00002-of-00010.safetensors",
+    "transformer.h.7.attn.c_attn.bias": "model-00003-of-00010.safetensors",
+    "transformer.h.7.attn.c_attn.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.7.attn.c_proj.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.7.ln_1.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.7.ln_2.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.7.mlp.c_proj.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.7.mlp.w1.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.7.mlp.w2.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.8.attn.c_attn.bias": "model-00003-of-00010.safetensors",
+    "transformer.h.8.attn.c_attn.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.8.attn.c_proj.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.8.ln_1.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.8.ln_2.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.8.mlp.c_proj.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.8.mlp.w1.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.8.mlp.w2.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.9.attn.c_attn.bias": "model-00003-of-00010.safetensors",
+    "transformer.h.9.attn.c_attn.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.9.attn.c_proj.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.9.ln_1.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.9.ln_2.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.9.mlp.c_proj.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.9.mlp.w1.weight": "model-00003-of-00010.safetensors",
+    "transformer.h.9.mlp.w2.weight": "model-00003-of-00010.safetensors",
+    "transformer.ln_f.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.attn_pool.attn.in_proj_bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.attn_pool.attn.in_proj_weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.attn_pool.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.attn_pool.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.attn_pool.kv_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.attn_pool.ln_kv.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.attn_pool.ln_kv.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.attn_pool.ln_q.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.attn_pool.ln_q.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.attn_pool.pos_embed": "model-00009-of-00010.safetensors",
+    "transformer.visual.attn_pool.query": "model-00009-of-00010.safetensors",
+    "transformer.visual.conv1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.ln_post.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.ln_post.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.ln_pre.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.ln_pre.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.positional_embedding": "model-00008-of-00010.safetensors",
+    "transformer.visual.proj": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.0.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.0.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.0.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.0.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.0.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.0.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.0.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.0.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.0.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.0.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.0.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.0.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.1.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.1.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.1.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.1.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.1.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.1.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.1.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.1.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.1.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.1.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.1.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.1.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.10.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.10.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.10.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.10.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.10.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.10.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.10.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.10.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.10.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.10.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.10.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.10.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.11.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.11.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.11.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.11.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.11.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.11.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.11.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.11.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.11.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.11.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.11.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.11.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.12.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.12.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.12.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.12.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.12.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.12.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.12.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.12.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.12.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.12.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.12.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.12.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.13.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.13.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.13.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.13.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.13.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.13.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.13.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.13.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.13.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.13.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.13.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.13.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.14.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.14.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.14.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.14.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.14.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.14.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.14.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.14.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.14.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.14.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.14.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.14.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.15.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.15.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.15.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.15.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.15.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.15.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.15.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.15.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.15.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.15.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.15.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.15.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.16.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.16.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.16.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.16.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.16.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.16.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.16.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.16.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.16.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.16.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.16.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.16.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.17.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.17.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.17.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.17.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.17.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.17.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.17.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.17.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.17.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.17.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.17.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.17.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.18.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.18.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.18.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.18.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.18.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.18.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.18.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.18.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.18.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.18.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.18.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.18.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.19.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.19.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.19.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.19.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.19.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.19.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.19.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.19.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.19.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.19.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.19.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.19.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.2.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.2.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.2.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.2.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.2.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.2.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.2.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.2.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.2.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.2.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.2.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.2.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.20.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.20.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.20.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.20.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.20.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.20.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.20.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.20.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.20.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.20.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.20.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.20.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.21.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.21.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.21.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.21.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.21.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.21.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.21.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.21.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.21.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.21.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.21.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.21.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.22.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.22.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.22.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.22.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.22.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.22.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.22.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.22.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.22.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.22.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.22.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.22.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.23.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.23.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.23.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.23.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.23.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.23.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.23.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.23.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.23.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.23.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.23.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.23.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.24.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.24.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.24.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.24.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.24.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.24.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.24.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.24.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.24.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.24.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.24.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.24.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.25.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.25.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.25.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.25.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.25.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.25.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.25.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.25.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.25.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.25.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.25.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.25.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.26.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.26.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.26.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.26.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.26.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.26.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.26.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.26.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.26.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.26.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.26.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.26.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.27.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.27.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.27.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.27.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.27.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.27.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.27.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.27.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.27.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.27.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.27.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.27.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.28.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.28.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.28.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.28.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.28.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.28.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.28.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.28.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.28.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.28.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.28.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.28.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.29.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.29.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.29.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.29.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.29.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.29.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.29.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.29.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.29.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.29.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.29.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.29.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.3.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.3.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.3.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.3.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.3.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.3.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.3.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.3.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.3.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.3.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.3.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.3.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.30.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.30.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.30.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.30.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.30.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.30.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.30.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.30.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.30.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.30.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.30.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.30.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.31.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.31.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.31.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.31.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.31.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.31.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.31.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.31.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.31.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.31.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.31.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.31.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.32.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.32.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.32.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.32.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.32.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.32.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.32.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.32.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.32.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.32.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.32.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.32.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.33.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.33.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.33.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.33.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.33.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.33.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.33.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.33.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.33.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.33.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.33.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.33.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.34.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.34.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.34.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.34.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.34.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.34.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.34.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.34.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.34.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.34.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.34.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.34.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.35.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.35.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.35.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.35.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.35.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.35.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.35.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.35.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.35.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.35.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.35.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.35.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.36.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.36.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.36.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.36.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.36.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.36.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.36.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.36.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.36.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.36.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.36.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.36.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.37.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.37.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.37.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.37.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.37.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.37.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.37.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.37.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.37.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.37.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.37.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.37.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.38.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.38.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.38.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.38.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.38.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.38.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.38.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.38.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.38.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.38.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.38.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.38.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.39.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.39.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.39.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.39.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.39.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.39.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.39.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.39.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.39.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.39.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.39.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.39.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.4.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.4.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.4.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.4.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.4.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.4.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.4.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.4.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.4.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.4.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.4.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.4.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.40.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.40.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.40.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.40.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.40.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.40.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.40.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.40.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.40.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.40.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.40.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.40.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.41.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.41.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.41.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.41.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.41.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.41.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.41.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.41.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.41.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.41.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.41.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.41.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.42.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.42.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.42.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.42.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.42.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.42.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.42.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.42.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.42.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.42.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.42.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.42.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.43.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.43.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.43.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.43.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.43.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.43.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.43.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.43.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.43.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.43.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.43.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.43.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.44.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.44.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.44.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.44.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.44.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.44.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.44.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.44.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.44.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.44.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.44.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.44.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.45.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.45.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.45.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.45.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.45.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.45.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.45.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.45.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.45.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.45.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.45.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.45.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.46.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.46.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.46.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.46.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.46.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.46.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.46.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.46.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.46.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.46.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.46.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.46.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.47.attn.in_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.47.attn.in_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.47.attn.out_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.47.attn.out_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.47.ln_1.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.47.ln_1.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.47.ln_2.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.47.ln_2.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.47.mlp.c_fc.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.47.mlp.c_fc.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.47.mlp.c_proj.bias": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.47.mlp.c_proj.weight": "model-00009-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.5.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.5.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.5.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.5.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.5.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.5.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.5.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.5.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.5.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.5.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.5.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.5.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.6.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.6.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.6.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.6.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.6.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.6.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.6.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.6.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.6.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.6.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.6.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.6.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.7.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.7.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.7.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.7.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.7.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.7.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.7.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.7.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.7.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.7.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.7.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.7.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.8.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.8.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.8.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.8.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.8.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.8.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.8.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.8.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.8.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.8.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.8.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.8.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.9.attn.in_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.9.attn.in_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.9.attn.out_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.9.attn.out_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.9.ln_1.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.9.ln_1.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.9.ln_2.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.9.ln_2.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.9.mlp.c_fc.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.9.mlp.c_fc.weight": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.9.mlp.c_proj.bias": "model-00008-of-00010.safetensors",
+    "transformer.visual.transformer.resblocks.9.mlp.c_proj.weight": "model-00008-of-00010.safetensors",
+    "transformer.wte.weight": "model-00001-of-00010.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1 @@
+{}

tokenization_qwen.py

@@ -0,0 +1,598 @@
+# Copyright (c) Alibaba Cloud.
+#
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+"""Tokenization classes for QWen."""
+
+import base64
+import logging
+import os
+import requests
+import unicodedata
+from typing import Collection, Dict, List, Set, Tuple, Union, Any, Callable, Optional
+
+import tiktoken
+import numpy as np
+from PIL import Image
+from PIL import ImageFont
+from PIL import ImageDraw
+from transformers import PreTrainedTokenizer, AddedToken
+from transformers.utils import try_to_load_from_cache
+
+import matplotlib.colors as mcolors
+from matplotlib.font_manager import FontProperties
+
+logger = logging.getLogger(__name__)
+
+
+VOCAB_FILES_NAMES = {"vocab_file": "qwen.tiktoken", "ttf": "SimSun.ttf"}
+FONT_PATH = try_to_load_from_cache("Qwen/Qwen-VL-Chat", "SimSun.ttf")
+if FONT_PATH is None:
+    if not os.path.exists("SimSun.ttf"):
+        ttf = requests.get("https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/SimSun.ttf")
+        open("SimSun.ttf", "wb").write(ttf.content)
+    FONT_PATH = "SimSun.ttf"
+
+PAT_STR = r"""(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\r\n\p{L}\p{N}]?\p{L}+|\p{N}| ?[^\s\p{L}\p{N}]+[\r\n]*|\s*[\r\n]+|\s+(?!\S)|\s+"""
+ENDOFTEXT = "<|endoftext|>"
+IMSTART = "<|im_start|>"
+IMEND = "<|im_end|>"
+# as the default behavior is changed to allow special tokens in
+# regular texts, the surface forms of special tokens need to be
+# as different as possible to minimize the impact
+EXTRAS = tuple((f"<|extra_{i}|>" for i in range(205)))
+SPECIAL_TOKENS = (
+    ENDOFTEXT,
+    IMSTART,
+    IMEND,
+) + EXTRAS
+IMG_TOKEN_SPAN = 256
+
+
+def _load_tiktoken_bpe(tiktoken_bpe_file: str) -> Dict[bytes, int]:
+    with open(tiktoken_bpe_file, "rb") as f:
+        contents = f.read()
+    return {
+        base64.b64decode(token): int(rank)
+        for token, rank in (line.split() for line in contents.splitlines() if line)
+    }
+
+def _list_find(
+    input_list: List[Any],
+    candidates: Tuple[Any],
+    start: int = 0,
+):
+    for i in range(start, len(input_list)):
+        if input_list[i] in candidates:
+            return i
+    return -1
+
+def _replace_closed_tag(
+    input_tokens: List[Any],
+    start_tags: Union[Any, Tuple[Any]],
+    end_tags: Union[Any, Tuple[Any]],
+    inclusive_replace_func: Callable,
+    exclusive_replace_func: Callable = lambda x: x,
+):
+    if isinstance(start_tags, (str, int)):
+        start_tags = (start_tags,)
+    if isinstance(end_tags, (str, int)):
+        end_tags = (end_tags,)
+    assert len(start_tags) == len(end_tags)
+
+    output_tokens = []
+    end = 0
+    while True:
+        start = _list_find(input_tokens, start_tags, end)
+        if start == -1:
+            break
+        output_tokens.extend(exclusive_replace_func(input_tokens[end : start]))
+        tag_idx = start_tags.index(input_tokens[start])
+        end = _list_find(input_tokens, (end_tags[tag_idx],), start)
+        if end == -1:
+            raise ValueError("Unclosed image token")
+        output_tokens.extend(inclusive_replace_func(input_tokens[start : end + 1]))
+        end += 1
+    output_tokens.extend(exclusive_replace_func(input_tokens[end : ]))
+    return output_tokens
+
+class QWenTokenizer(PreTrainedTokenizer):
+    """QWen tokenizer."""
+
+    vocab_files_names = VOCAB_FILES_NAMES
+
+    def __init__(
+        self,
+        vocab_file,
+        errors="replace",
+        image_start_tag='<img>',
+        image_end_tag='</img>',
+        image_pad_tag='<imgpad>',
+        ref_start_tag='<ref>',
+        ref_end_tag='</ref>',
+        box_start_tag='<box>',
+        box_end_tag='</box>',
+        quad_start_tag='<quad>',
+        quad_end_tag='</quad>',
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+        self.image_start_tag = image_start_tag
+        self.image_end_tag = image_end_tag
+        self.image_pad_tag = image_pad_tag
+        self.ref_start_tag = ref_start_tag
+        self.ref_end_tag = ref_end_tag
+        self.box_start_tag = box_start_tag
+        self.box_end_tag = box_end_tag
+        self.quad_start_tag = quad_start_tag
+        self.quad_end_tag = quad_end_tag
+        self.IMAGE_ST = (
+            ref_start_tag, ref_end_tag,
+            box_start_tag, box_end_tag,
+            quad_start_tag, quad_end_tag,
+            image_start_tag, image_end_tag,
+            image_pad_tag
+        )
+
+        self.errors = errors  # how to handle errors in decoding
+
+        self.mergeable_ranks = _load_tiktoken_bpe(vocab_file)  # type: dict[bytes, int]
+        self.special_tokens = {
+            token: index
+            for index, token in enumerate(
+                SPECIAL_TOKENS + self.IMAGE_ST, start=len(self.mergeable_ranks)
+            )
+        }
+        self.img_start_id = self.special_tokens[self.image_start_tag]
+        self.img_end_id = self.special_tokens[self.image_end_tag]
+        self.img_pad_id = self.special_tokens[self.image_pad_tag]
+        self.ref_start_id = self.special_tokens[self.ref_start_tag]
+        self.ref_end_id = self.special_tokens[self.ref_end_tag]
+        self.box_start_id = self.special_tokens[self.box_start_tag]
+        self.box_end_id = self.special_tokens[self.box_end_tag]
+        self.quad_start_id = self.special_tokens[self.quad_start_tag]
+        self.quad_end_id = self.special_tokens[self.quad_end_tag]
+        self.image_special_tokens = set([
+            self.ref_start_id, self.ref_end_id, self.box_start_id, self.box_end_id,
+            self.quad_start_id, self.quad_end_id,
+        ])
+
+        enc = tiktoken.Encoding(
+            "Qwen",
+            pat_str=PAT_STR,
+            mergeable_ranks=self.mergeable_ranks,
+            special_tokens=self.special_tokens,
+        )
+        assert (
+            len(self.mergeable_ranks) + len(self.special_tokens) == enc.n_vocab
+        ), f"{len(self.mergeable_ranks) + len(self.special_tokens)} != {enc.n_vocab} in encoding"
+
+        self.decoder = {
+            v: k for k, v in self.mergeable_ranks.items()
+        }  # type: dict[int, bytes|str]
+        self.decoder.update({v: k for k, v in self.special_tokens.items()})
+
+        self.tokenizer = enc  # type: tiktoken.Encoding
+
+        self.eod_id = self.tokenizer.eot_token
+        self.im_start_id = self.special_tokens[IMSTART]
+        self.im_end_id = self.special_tokens[IMEND]
+
+    def __getstate__(self):
+        # for pickle lovers
+        state = self.__dict__.copy()
+        del state['tokenizer']
+        return state
+
+    def __setstate__(self, state):
+        # tokenizer is not python native; don't pass it; rebuild it
+        self.__dict__.update(state)
+        enc = tiktoken.Encoding(
+            "Qwen",
+            pat_str=PAT_STR,
+            mergeable_ranks=self.mergeable_ranks,
+            special_tokens=self.special_tokens,
+        )
+        self.tokenizer = enc
+
+
+    def __len__(self) -> int:
+        return self.tokenizer.n_vocab
+
+    def get_vocab(self) -> Dict[bytes, int]:
+        return self.mergeable_ranks
+
+    def convert_tokens_to_ids(
+        self, tokens: Union[bytes, str, List[Union[bytes, str]]]
+    ) -> List[int]:
+        ids = []
+        if isinstance(tokens, (str, bytes)):
+            if tokens in self.special_tokens:
+                return self.special_tokens[tokens]
+            else:
+                return self.mergeable_ranks.get(tokens)
+        for token in tokens:
+            if token in self.special_tokens:
+                ids.append(self.special_tokens[token])
+            else:
+                ids.append(self.mergeable_ranks.get(token))
+        return ids
+
+    def _add_tokens(self, new_tokens: Union[List[str], List[AddedToken]], special_tokens: bool = False) -> int:
+        if not special_tokens and new_tokens:
+            raise ValueError('Adding regular tokens is not supported')
+        for token in new_tokens:
+            surface_form = token.content if isinstance(token, AddedToken) else token
+            if surface_form not in SPECIAL_TOKENS + self.IMAGE_ST:
+                raise ValueError('Adding unknown special tokens is not supported')
+        return 0
+
+    def save_vocabulary(self, save_directory: str, **kwargs) -> Tuple[str]:
+        """
+        Save only the vocabulary of the tokenizer (vocabulary).
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        file_path = os.path.join(save_directory, "qwen.tiktoken")
+        with open(file_path, "w", encoding="utf8") as w:
+            for k, v in self.mergeable_ranks.items():
+                line = base64.b64encode(k).decode("utf8") + " " + str(v) + "\n"
+                w.write(line)
+        return (file_path,)
+
+    def tokenize(
+        self,
+        text: str,
+        allowed_special: Union[Set, str] = "all",
+        disallowed_special: Union[Collection, str] = (),
+        **kwargs,
+    ) -> List[Union[bytes, str]]:
+        """
+        Converts a string in a sequence of tokens.
+
+        Args:
+            text (`str`):
+                The sequence to be encoded.
+            allowed_special (`Literal["all"]` or `set`):
+                The surface forms of the tokens to be encoded as special tokens in regular texts.
+                Default to "all".
+            disallowed_special (`Literal["all"]` or `Collection`):
+                The surface forms of the tokens that should not be in regular texts and trigger errors.
+                Default to an empty tuple.
+
+            kwargs (additional keyword arguments, *optional*):
+                Will be passed to the underlying model specific encode method.
+
+        Returns:
+            `List[bytes|str]`: The list of tokens.
+        """
+        tokens = []
+        text = unicodedata.normalize("NFC", text)
+
+        # this implementation takes a detour: text -> token id -> token surface forms
+        for t in self.tokenizer.encode(
+            text, allowed_special=allowed_special, disallowed_special=disallowed_special
+        ):
+            tokens.append(self.decoder[t])
+
+        def _encode_imgurl(img_tokens):
+            assert img_tokens[0] == self.image_start_tag and img_tokens[-1] == self.image_end_tag
+            img_tokens = img_tokens[1:-1]
+            img_url = b''.join(img_tokens)
+            out_img_tokens = list(map(self.decoder.get, img_url))
+            if len(out_img_tokens) > IMG_TOKEN_SPAN:
+                raise ValueError("The content in {}..{} is too long".format(
+                    self.image_start_tag, self.image_end_tag))
+            out_img_tokens.extend([self.image_pad_tag] * (IMG_TOKEN_SPAN - len(out_img_tokens)))
+            out_img_tokens = [self.image_start_tag] + out_img_tokens + [self.image_end_tag]
+            return out_img_tokens
+
+        return _replace_closed_tag(tokens, self.image_start_tag, self.image_end_tag, _encode_imgurl)
+
+    def convert_tokens_to_string(self, tokens: List[Union[bytes, str]]) -> str:
+        """
+        Converts a sequence of tokens in a single string.
+        """
+        text = ""
+        temp = b""
+        for t in tokens:
+            if isinstance(t, str):
+                if temp:
+                    text += temp.decode("utf-8", errors=self.errors)
+                    temp = b""
+                text += t
+            elif isinstance(t, bytes):
+                temp += t
+            else:
+                raise TypeError("token should only be of type types or str")
+        if temp:
+            text += temp.decode("utf-8", errors=self.errors)
+        return text
+
+    @property
+    def vocab_size(self):
+        return self.tokenizer.n_vocab
+
+    def _convert_id_to_token(self, index: int) -> Union[bytes, str]:
+        """Converts an id to a token, special tokens included"""
+        if index in self.decoder:
+            return self.decoder[index]
+        raise ValueError("unknown ids")
+
+    def _convert_token_to_id(self, token: Union[bytes, str]) -> int:
+        """Converts a token to an id using the vocab, special tokens included"""
+        if token in self.special_tokens:
+            return self.special_tokens[token]
+        if token in self.mergeable_ranks:
+            return self.mergeable_ranks[token]
+        raise ValueError("unknown token")
+
+    def _tokenize(self, text: str, **kwargs):
+        """
+        Converts a string in a sequence of tokens (string), using the tokenizer. Split in words for word-based
+        vocabulary or sub-words for sub-word-based vocabularies (BPE/SentencePieces/WordPieces).
+
+        Do NOT take care of added tokens.
+        """
+        raise NotImplementedError
+
+    def _decode(
+        self,
+        token_ids: Union[int, List[int]],
+        skip_special_tokens: bool = False,
+        errors: str = None,
+        **kwargs,
+    ) -> str:
+        if isinstance(token_ids, int):
+            token_ids = [token_ids]
+
+        def _decode_imgurl(img_token_ids):
+            assert img_token_ids[0] == self.img_start_id and img_token_ids[-1] == self.img_end_id
+            img_token_ids = img_token_ids[1:-1]
+            img_token_ids = img_token_ids[ : img_token_ids.index(self.img_pad_id)]
+            img_url = bytes(img_token_ids).decode('utf-8')
+            return [self.img_start_id] + self.tokenizer.encode(img_url) + [self.img_end_id]
+
+        token_ids = _replace_closed_tag(token_ids, self.img_start_id, self.img_end_id, _decode_imgurl)
+
+        if skip_special_tokens:
+            if kwargs.get('keep_image_special', False):
+                token_ids = [i for i in token_ids if i < self.eod_id
+                    or i in self.image_special_tokens]
+            else:
+                token_ids = [i for i in token_ids if i < self.eod_id]
+        return self.tokenizer.decode(token_ids, errors=errors or self.errors)
+
+    def to_list_format(self, text: str):
+        text = unicodedata.normalize("NFC", text)
+        token_ids = self.tokenizer.encode(
+            text, allowed_special=set(self.IMAGE_ST + (ENDOFTEXT,)))
+
+        def _encode_vl_info(tokens):
+            if len(tokens) == 0:
+                return []
+            if tokens[0] == self.img_start_id and tokens[-1] == self.img_end_id:
+                key = 'image'
+            elif tokens[0] == self.ref_start_id and tokens[-1] == self.ref_end_id:
+                key = 'ref'
+            elif tokens[0] == self.box_start_id and tokens[-1] == self.box_end_id:
+                key = 'box'
+            elif tokens[0] == self.quad_start_id and tokens[-1] == self.quad_end_id:
+                key = 'quad'
+            else:
+                _tobytes = lambda x: x.encode('utf-8') if isinstance(x, str) else x
+                return [{'text': b''.join(map(_tobytes, map(self.decoder.get, tokens))).decode('utf-8')}]
+            _tobytes = lambda x: x.encode('utf-8') if isinstance(x, str) else x
+            val = b''.join(map(_tobytes, map(self.decoder.get, tokens[1:-1]))).decode('utf-8')
+            return [{key: val}]
+
+        return _replace_closed_tag(
+            token_ids,
+            (self.img_start_id, self.ref_start_id, self.box_start_id, self.quad_start_id),
+            (self.img_end_id, self.ref_end_id, self.box_end_id, self.quad_end_id),
+            _encode_vl_info,
+            _encode_vl_info,
+        )
+
+    def from_list_format(self, list_format: List[Dict]):
+        text = ''
+        num_images = 0
+        for ele in list_format:
+            if 'image' in ele:
+                num_images += 1
+                text += f'Picture {num_images}: '
+                text += self.image_start_tag + ele['image'] + self.image_end_tag
+                text += '\n'
+            elif 'text' in ele:
+                text += ele['text']
+            elif 'box' in ele:
+                if 'ref' in ele:
+                    text += self.ref_start_tag + ele['ref'] + self.ref_end_tag
+                for box in ele['box']:
+                    text += self.box_start_tag + '(%d,%d),(%d,%d)' % (box[0], box[1], box[2], box[3]) + self.box_end_tag
+            else:
+                raise ValueError("Unsupport element: " + str(ele))
+        return text
+
+    def _fetch_latest_picture(self, response, history):
+        if history is None:
+            history = []
+        _history = history + [(response, None)]
+        for q, r in _history[::-1]:
+            for ele in self.to_list_format(q)[::-1]:
+                if 'image' in ele:
+                    return ele['image']
+        return None
+
+    def _fetch_all_box_with_ref(self, text):
+        list_format = self.to_list_format(text)
+        output = []
+        for i, ele in enumerate(list_format):
+            if 'box' in ele:
+                bbox = tuple(map(int, ele['box'].replace('(', '').replace(')', '').split(',')))
+                assert len(bbox) == 4
+                output.append({'box': bbox})
+                if i > 0 and 'ref' in list_format[i-1]:
+                    output[-1]['ref'] = list_format[i-1]['ref'].strip()
+        return output
+
+    def draw_bbox_on_latest_picture(
+        self,
+        response,
+        history=None,
+    ) -> Optional[Image.Image]:
+        image = self._fetch_latest_picture(response, history)
+        if image is None:
+            return None
+        if image.startswith("http://") or image.startswith("https://"):
+            image = Image.open(requests.get(image, stream=True).raw).convert("RGB")
+            h, w = image.height, image.width
+        else:
+            image = np.asarray(Image.open(image).convert("RGB"))
+            h, w = image.shape[0], image.shape[1]
+        visualizer = Visualizer(image)
+
+        boxes = self._fetch_all_box_with_ref(response)
+        if not boxes:
+            return None
+        color = random.choice([_ for _ in mcolors.TABLEAU_COLORS.keys()]) # init color
+        for box in boxes:
+            if 'ref' in box: # random new color for new refexps
+                color = random.choice([_ for _ in mcolors.TABLEAU_COLORS.keys()])
+            x1, y1, x2, y2 = box['box']
+            x1, y1, x2, y2 = (int(x1 / 1000 * w), int(y1 / 1000 * h), int(x2 / 1000 * w), int(y2 / 1000 * h))
+            visualizer.draw_box((x1, y1, x2, y2), alpha=1, edge_color=color)
+            if 'ref' in box:
+                visualizer.draw_text(box['ref'], (x1, y1), color=color, horizontal_alignment="left")
+        return visualizer.output
+
+
+import colorsys
+import logging
+import math
+import numpy as np
+import matplotlib as mpl
+import matplotlib.colors as mplc
+import matplotlib.figure as mplfigure
+import torch
+from matplotlib.backends.backend_agg import FigureCanvasAgg
+from PIL import Image
+import random
+
+logger = logging.getLogger(__name__)
+
+
+class VisImage:
+    def __init__(self, img, scale=1.0):
+        self.img = img
+        self.scale = scale
+        self.width, self.height = img.shape[1], img.shape[0]
+        self._setup_figure(img)
+
+    def _setup_figure(self, img):
+        fig = mplfigure.Figure(frameon=False)
+        self.dpi = fig.get_dpi()
+        # add a small 1e-2 to avoid precision lost due to matplotlib's truncation
+        # (https://github.com/matplotlib/matplotlib/issues/15363)
+        fig.set_size_inches(
+            (self.width * self.scale + 1e-2) / self.dpi,
+            (self.height * self.scale + 1e-2) / self.dpi,
+        )
+        self.canvas = FigureCanvasAgg(fig)
+        # self.canvas = mpl.backends.backend_cairo.FigureCanvasCairo(fig)
+        ax = fig.add_axes([0.0, 0.0, 1.0, 1.0])
+        ax.axis("off")
+        self.fig = fig
+        self.ax = ax
+        self.reset_image(img)
+
+    def reset_image(self, img):
+        img = img.astype("uint8")
+        self.ax.imshow(img, extent=(0, self.width, self.height, 0), interpolation="nearest")
+
+    def save(self, filepath):
+        self.fig.savefig(filepath)
+
+    def get_image(self):
+        canvas = self.canvas
+        s, (width, height) = canvas.print_to_buffer()
+
+        buffer = np.frombuffer(s, dtype="uint8")
+
+        img_rgba = buffer.reshape(height, width, 4)
+        rgb, alpha = np.split(img_rgba, [3], axis=2)
+        return rgb.astype("uint8")
+
+
+class Visualizer:
+    def __init__(self, img_rgb, metadata=None, scale=1.0):
+        self.img = np.asarray(img_rgb).clip(0, 255).astype(np.uint8)
+        self.font_path = FONT_PATH
+        self.output = VisImage(self.img, scale=scale)
+        self.cpu_device = torch.device("cpu")
+
+        # too small texts are useless, therefore clamp to 14
+        self._default_font_size = max(
+            np.sqrt(self.output.height * self.output.width) // 30, 15 // scale
+        )
+
+    def draw_text(
+        self,
+        text,
+        position,
+        *,
+        font_size=None,
+        color="g",
+        horizontal_alignment="center",
+        rotation=0,
+    ):
+        if not font_size:
+            font_size = self._default_font_size
+
+        # since the text background is dark, we don't want the text to be dark
+        color = np.maximum(list(mplc.to_rgb(color)), 0.2)
+        color[np.argmax(color)] = max(0.8, np.max(color))
+
+        x, y = position
+        self.output.ax.text(
+            x,
+            y,
+            text,
+            size=font_size * self.output.scale,
+            fontproperties=FontProperties(fname=self.font_path),
+            bbox={"facecolor": "black", "alpha": 0.8, "pad": 0.7, "edgecolor": "none"},
+            verticalalignment="top",
+            horizontalalignment=horizontal_alignment,
+            color=color,
+            zorder=10,
+            rotation=rotation,
+        )
+        return self.output
+
+    def draw_box(self, box_coord, alpha=0.5, edge_color="g", line_style="-"):
+        
+        x0, y0, x1, y1 = box_coord
+        width = x1 - x0
+        height = y1 - y0
+
+        linewidth = max(self._default_font_size / 4, 1)
+
+        self.output.ax.add_patch(
+            mpl.patches.Rectangle(
+                (x0, y0),
+                width,
+                height,
+                fill=False,
+                edgecolor=edge_color,
+                linewidth=linewidth * self.output.scale,
+                alpha=alpha,
+                linestyle=line_style,
+            )
+        )
+        return self.output
+
+    def get_output(self):
+        
+        return self.output

tokenizer_config.json

@@ -0,0 +1,11 @@
+{
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_qwen.QWenTokenizer",
+      null
+    ]
+  },
+  "clean_up_tokenization_spaces": true,
+  "model_max_length": 8192,
+  "tokenizer_class": "QWenTokenizer"
+}