push merged model
#4
by
Adirazgold
- opened
- .gitattributes +0 -1
- README.md +0 -140
- REPORT_Benchmarking the AI advantage in finance.pdf +0 -3
- added_tokens.json +0 -6
- chat_template.json +0 -3
- config.json +0 -179
- granite_vision_embedding_config.py +0 -15
- merges.txt +0 -0
- model-00001-of-00003.safetensors +0 -3
- model-00002-of-00003.safetensors +0 -3
- model-00003-of-00003.safetensors +0 -3
- model.safetensors.index.json +0 -824
- modeling_granite_vision_embedding.py +0 -190
- preprocessor_config.json +0 -137
- processing_granite_vision_embedding.py +0 -439
- processor_config.json +0 -6
- special_tokens_map.json +0 -35
- tokenizer.json +0 -0
- tokenizer_config.json +0 -208
- vocab.json +0 -0
.gitattributes
CHANGED
|
@@ -33,4 +33,3 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
-
REPORT_Benchmarking[[:space:]]the[[:space:]]AI[[:space:]]advantage[[:space:]]in[[:space:]]finance.pdf filter=lfs diff=lfs merge=lfs -text
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
README.md
DELETED
|
@@ -1,140 +0,0 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
language:
|
| 4 |
-
- en
|
| 5 |
-
base_model:
|
| 6 |
-
- ibm-granite/granite-vision-3.3-2b
|
| 7 |
-
library_name: transformers
|
| 8 |
-
---
|
| 9 |
-
# granite-vision-3.3-2b-embedding
|
| 10 |
-
**Model Summary:**
|
| 11 |
-
Granite-vision-3.3-2b-embedding is an efficient embedding model based on [granite-vision-3.3-2b](https://huggingface.co/ibm-granite/granite-vision-3.3-2b). This model is specifically designed for multimodal document retrieval, enabling queries on documents with tables, charts, infographics, and complex layouts. The model generates ColBERT-style multi-vector representations of pages.
|
| 12 |
-
By removing the need for OCR-based text extractions, granite-vision-3.3-2b-embedding can help simplify and accelerate RAG pipelines.
|
| 13 |
-
|
| 14 |
-
**Evaluations:**
|
| 15 |
-
We evaluated granite-vision-3.3-2b-embedding alongside other top colBERT style multi-modal embedding models in the 1B-4B parameter range using two benchmark: [Vidore2](https://github.com/illuin-tech/vidore-benchmark/) and [Real-MM-RAG-Bench](https://huggingface.co/collections/ibm-research/real-mm-rag-bench-67d2dc0ddf2dfafe66f09d34) which aim to specifically address complex multimodal document retrieval tasks.
|
| 16 |
-
|
| 17 |
-
## **NDCG@5 - ViDoRe V2**
|
| 18 |
-
| Collection \ Model | ColPali-v1.3 | ColQwen2.5-v0.2 | ColNomic-3b | ColSmolvlm-v0.1 | granite-vision-3.3-2b-embedding |
|
| 19 |
-
|----------------------------------------|--------------|------------------|-------------|-------------------|-----------
|
| 20 |
-
| ESG Restaurant Human | 51.1 | 68.4 | 65.8 | 62.4 | 65.3 |
|
| 21 |
-
| Economics Macro Multilingual | 49.9 | 56.5 | 55.4 | 47.4 | 51.2 |
|
| 22 |
-
| MIT Biomedical | 59.7 | 63.6 | 63.5 | 58.1 |61.5 |
|
| 23 |
-
| ESG Restaurant Synthetic | 57.0 | 57.4 | 56.6 | 51.1 |56.6 |
|
| 24 |
-
| ESG Restaurant Synthetic Multilingual | 55.7 | 57.4 | 57.2 | 47.6 |55.7 |
|
| 25 |
-
| MIT Biomedical Multilingual | 56.5 | 61.1 | 62.5 | 50.5 | 55.5 |
|
| 26 |
-
| Economics Macro | 51.6 | 59.8 | 60.2 | 60.9 |58.3 |
|
| 27 |
-
| **Avg (ViDoRe2)** | **54.5** | **60.6** | **60.2** | **54.0** |**57.7** |
|
| 28 |
-
|
| 29 |
-
## **NDCG@5 - REAL-MM-RAG**
|
| 30 |
-
| Collection \ Model | ColPali-v1.3 | ColQwen2.5-v0.2 | ColNomic-3b | ColSmolvlm-v0.1 | granite-vision-3.3-2b-embedding |
|
| 31 |
-
|----------------------------------------|--------------|------------------|-------------|--------------------------| ------------------
|
| 32 |
-
| FinReport | 55 | 66 | 78 | 65 |73
|
| 33 |
-
| FinSlides | 68 | 79 | 81 | 55 |79
|
| 34 |
-
| TechReport | 78 | 86 | 88 | 83 |87
|
| 35 |
-
| TechSlides | 90 | 93 | 92 | 91 |93
|
| 36 |
-
| **Avg (REAL-MM-RAG)** | **73** | **81** | **85** | **74** |**83**
|
| 37 |
-
|
| 38 |
-
- **Release Date**: June 11th 2025
|
| 39 |
-
- **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)
|
| 40 |
-
- **Supported Input Format:** Currently the model supports English instructions and images (png, jpeg) as input format.
|
| 41 |
-
|
| 42 |
-
**Intended Use:**
|
| 43 |
-
The model is intended to be used in enterprise applications that involve retrieval of visual and text data. In particular, the model is well-suited for multi-modal RAG systems where the knowledge base is composed of complex enterprise documents, such as reports, slides, images, canned doscuments, manuals and more. The model can be used as a standalone retriever, or alongside a text-based retriever.
|
| 44 |
-
|
| 45 |
-
### Usage
|
| 46 |
-
```shell
|
| 47 |
-
pip install -q torch torchvision torchaudio
|
| 48 |
-
pip install transformers==4.50
|
| 49 |
-
```
|
| 50 |
-
Then run the code:
|
| 51 |
-
```python
|
| 52 |
-
from io import BytesIO
|
| 53 |
-
|
| 54 |
-
import requests
|
| 55 |
-
import torch
|
| 56 |
-
from PIL import Image
|
| 57 |
-
from transformers import AutoProcessor, AutoModel
|
| 58 |
-
from transformers.utils.import_utils import is_flash_attn_2_available
|
| 59 |
-
|
| 60 |
-
device = "cuda" if torch.cuda.is_available() else "cpu"
|
| 61 |
-
model_name = "ibm-granite/granite-vision-3.3-2b-embedding"
|
| 62 |
-
model = AutoModel.from_pretrained(
|
| 63 |
-
model_name,
|
| 64 |
-
trust_remote_code=True,
|
| 65 |
-
torch_dtype=torch.float16,
|
| 66 |
-
device_map=device,
|
| 67 |
-
attn_implementation="flash_attention_2" if is_flash_attn_2_available() else None
|
| 68 |
-
).eval()
|
| 69 |
-
processor = AutoProcessor.from_pretrained(model_name, trust_remote_code=True)
|
| 70 |
-
|
| 71 |
-
# ─────────────────────────────────────────���───
|
| 72 |
-
# Inputs: Image + Text
|
| 73 |
-
# ─────────────────────────────────────────────
|
| 74 |
-
image_url = "https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg"
|
| 75 |
-
print("\nFetching image...")
|
| 76 |
-
image = Image.open(BytesIO(requests.get(image_url).content)).convert("RGB")
|
| 77 |
-
|
| 78 |
-
text = "A photo of a tiger"
|
| 79 |
-
print(f"Image and text inputs ready.")
|
| 80 |
-
|
| 81 |
-
# Process both inputs
|
| 82 |
-
print("Processing inputs...")
|
| 83 |
-
image_inputs = processor.process_images([image])
|
| 84 |
-
text_inputs = processor.process_queries([text])
|
| 85 |
-
|
| 86 |
-
# Move to correct device
|
| 87 |
-
image_inputs = {k: v.to(device) for k, v in image_inputs.items()}
|
| 88 |
-
text_inputs = {k: v.to(device) for k, v in text_inputs.items()}
|
| 89 |
-
|
| 90 |
-
# ─────────────────────────────────────────────
|
| 91 |
-
# Run Inference
|
| 92 |
-
# ─────────────────────────────────────────────
|
| 93 |
-
with torch.no_grad():
|
| 94 |
-
print("🔍 Getting image embedding...")
|
| 95 |
-
img_emb = model(**image_inputs)
|
| 96 |
-
|
| 97 |
-
print("✍️ Getting text embedding...")
|
| 98 |
-
txt_emb = model(**text_inputs)
|
| 99 |
-
|
| 100 |
-
# ─────────────────────────────────────────────
|
| 101 |
-
# Score the similarity
|
| 102 |
-
# ─────────────────────────────────────────────
|
| 103 |
-
print("Scoring similarity...")
|
| 104 |
-
similarity = processor.score(txt_emb, img_emb, batch_size=1, device=device)
|
| 105 |
-
|
| 106 |
-
print("\n" + "=" * 50)
|
| 107 |
-
print(f"📊 Similarity between image and text: {similarity.item():.4f}")
|
| 108 |
-
print("=" * 50)
|
| 109 |
-
```
|
| 110 |
-
### Use granite-vision-embedding-3.3-2b for MM RAG
|
| 111 |
-
For an example of MM-RAG using granite-vision-3.3-2b-embedding refer to [this notebook](https://github.com/ibm-granite/granite-vision-models/blob/main/cookbooks/GraniteVisionEmbedding_MM-RAG_Notebook.ipynb).
|
| 112 |
-
|
| 113 |
-
**Model Architecture:**
|
| 114 |
-
The architecture of granite-vision-3.3-2b-embedding follows ColPali(https://arxiv.org/abs/2407.01449) approach and consists of the following components:
|
| 115 |
-
|
| 116 |
-
(1) Vision-Language model : granite-vision-3.3-2b (https://huggingface.co/ibm-granite/granite-vision-3.3-2b).
|
| 117 |
-
|
| 118 |
-
(2) Projection layer: linear layer that projects the hidden layer dimension of Vision-Language model to 128 and outputs 729 embedding vectors per image.
|
| 119 |
-
|
| 120 |
-
The scoring is computed using MaxSim-based late interaction mechanism.
|
| 121 |
-
|
| 122 |
-
**Training Data:**
|
| 123 |
-
Our training data is entirly comprised from DocFM. DocFM is a large-scale comprehensive dataset effort at IBM consisting of 85 million document pages extracted from unique PDF
|
| 124 |
-
documents sourced from Common Crawl, Wikipedia, and ESG (Environmental, Social, and Governance)
|
| 125 |
-
reports.
|
| 126 |
-
|
| 127 |
-
**Infrastructure:**
|
| 128 |
-
We train granite-vision-3.3-2b-embedding on IBM’s cognitive computing cluster, which is outfitted with NVIDIA A100 GPUs.
|
| 129 |
-
|
| 130 |
-
**Ethical Considerations and Limitations:**
|
| 131 |
-
The use of Large Vision and Language Models involves risks and ethical considerations people must be aware of, including but not limited to: bias and fairness, misinformation, and autonomous decision-making. Granite-vision-3.3-2b-embedding is not the exception in this regard. Although our alignment processes include safety considerations, the model may in some cases produce inaccurate or biased responses.
|
| 132 |
-
Regarding ethics, a latent risk associated with all Large Language Models is their malicious utilization. We urge the community to use granite-vision-3.3-2b-embedding with ethical intentions and in a responsible way.
|
| 133 |
-
|
| 134 |
-
**Resources**
|
| 135 |
-
- 📄 Granite Vision technical report [here](https://arxiv.org/abs/2502.09927)
|
| 136 |
-
- 📄 Real-MM-RAG-Bench paper (ACL 2025) [here](https://arxiv.org/abs/2502.12342)
|
| 137 |
-
- 📄 Vidore 2 paper [here](https://www.arxiv.org/pdf/2505.17166)
|
| 138 |
-
- ⭐️ Learn about the latest updates with Granite: https://www.ibm.com/granite
|
| 139 |
-
- 🚀 Get started with tutorials, best practices, and prompt engineering advice: https://www.ibm.com/granite/docs/
|
| 140 |
-
- 💡 Learn about the latest Granite learning resources: https://ibm.biz/granite-learning-resources
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
REPORT_Benchmarking the AI advantage in finance.pdf
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:4e6da951c55eef3fd52aa41543f3b4377ab26e2758c579aec2d11068a66b3d20
|
| 3 |
-
size 1746880
|
|
|
|
|
|
|
|
|
|
|
|
added_tokens.json
DELETED
|
@@ -1,6 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"<image>": 49155,
|
| 3 |
-
"<|end_of_role|>": 49153,
|
| 4 |
-
"<|start_of_role|>": 49152,
|
| 5 |
-
"<|tool_call|>": 49154
|
| 6 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
chat_template.json
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"chat_template": "{%- if tools %}\n {{- '<|start_of_role|>available_tools<|end_of_role|>\n' }}\n {%- for tool in tools %}\n {{- tool | tojson(indent=4) }}\n {%- if not loop.last %}\n {{- '\n\n' }}\n {%- endif %}\n {%- endfor %}\n {{- '<|end_of_text|>\n' }}\n{%- endif %}\n{%- for message in messages if message['role'] == 'system'%}{% else %}<|system|>\nA chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.\n{% endfor %}{%- for message in messages %}\n {%- if message['role'] == 'system' %}\n {{- '<|system|>\n' + message['content'][0]['text'] + '\n' }}\n {%- elif message['role'] == 'user' %}<|user|>\n {# Render all images first #}{% for content in message['content'] | selectattr('type', 'equalto', 'image') %}{{ '<image>\n' }}{% endfor %}{# Render all text next #}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{{ content['text'] + '\n' }}{% endfor %}\n{%- elif message['role'] == 'assistant' %}\n {{- '<|assistant|>\n' + message['content'][0]['text'] + '<|end_of_text|>' }}\n {%- elif message['role'] == 'assistant_tool_call' %}\n {{- '<|start_of_role|>assistant<|end_of_role|><|tool_call|>' + message['content'][0]['text'] + '<|end_of_text|>\n' }}\n {%- elif message['role'] == 'tool_response' %}\n {{- '<|start_of_role|>tool_response<|end_of_role|>' + message['content'][0]['text'] + '<|end_of_text|>\n' }}\n {%- endif %}\n {%- if loop.last and add_generation_prompt %}\n {{- '<|assistant|>\n' }}\n {%- endif %}\n{%- endfor %}"
|
| 3 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
config.json
DELETED
|
@@ -1,179 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"_name_or_path": "ibm_granite/granite-vision-3.3-2b",
|
| 3 |
-
"adapter_path": null,
|
| 4 |
-
"auto_map": {
|
| 5 |
-
"AutoModel": "modeling_granite_vision_embedding.GraniteVisionEmb",
|
| 6 |
-
"AutoProcessor": "processing_granite_vision_embedding.GraniteVisionEmbProcessor",
|
| 7 |
-
"AutoConfig": "granite_vision_embedding_config.GraniteVisionEmbConfig"
|
| 8 |
-
},
|
| 9 |
-
"architectures": [
|
| 10 |
-
"GraniteVisionEmb"
|
| 11 |
-
],
|
| 12 |
-
"base_image_feature_location": "last",
|
| 13 |
-
"base_model": null,
|
| 14 |
-
"emb_dim_doc": 128,
|
| 15 |
-
"emb_dim_query": 128,
|
| 16 |
-
"image_grid_pinpoints": [
|
| 17 |
-
[
|
| 18 |
-
384,
|
| 19 |
-
768
|
| 20 |
-
],
|
| 21 |
-
[
|
| 22 |
-
384,
|
| 23 |
-
1152
|
| 24 |
-
],
|
| 25 |
-
[
|
| 26 |
-
384,
|
| 27 |
-
1536
|
| 28 |
-
],
|
| 29 |
-
[
|
| 30 |
-
384,
|
| 31 |
-
1920
|
| 32 |
-
],
|
| 33 |
-
[
|
| 34 |
-
384,
|
| 35 |
-
2304
|
| 36 |
-
],
|
| 37 |
-
[
|
| 38 |
-
384,
|
| 39 |
-
2688
|
| 40 |
-
],
|
| 41 |
-
[
|
| 42 |
-
384,
|
| 43 |
-
3072
|
| 44 |
-
],
|
| 45 |
-
[
|
| 46 |
-
384,
|
| 47 |
-
3456
|
| 48 |
-
],
|
| 49 |
-
[
|
| 50 |
-
384,
|
| 51 |
-
3840
|
| 52 |
-
],
|
| 53 |
-
[
|
| 54 |
-
768,
|
| 55 |
-
384
|
| 56 |
-
],
|
| 57 |
-
[
|
| 58 |
-
768,
|
| 59 |
-
768
|
| 60 |
-
],
|
| 61 |
-
[
|
| 62 |
-
768,
|
| 63 |
-
1152
|
| 64 |
-
],
|
| 65 |
-
[
|
| 66 |
-
768,
|
| 67 |
-
1536
|
| 68 |
-
],
|
| 69 |
-
[
|
| 70 |
-
768,
|
| 71 |
-
1920
|
| 72 |
-
],
|
| 73 |
-
[
|
| 74 |
-
1152,
|
| 75 |
-
384
|
| 76 |
-
],
|
| 77 |
-
[
|
| 78 |
-
1152,
|
| 79 |
-
768
|
| 80 |
-
],
|
| 81 |
-
[
|
| 82 |
-
1152,
|
| 83 |
-
1152
|
| 84 |
-
],
|
| 85 |
-
[
|
| 86 |
-
1536,
|
| 87 |
-
384
|
| 88 |
-
],
|
| 89 |
-
[
|
| 90 |
-
1536,
|
| 91 |
-
768
|
| 92 |
-
],
|
| 93 |
-
[
|
| 94 |
-
1920,
|
| 95 |
-
384
|
| 96 |
-
],
|
| 97 |
-
[
|
| 98 |
-
1920,
|
| 99 |
-
768
|
| 100 |
-
],
|
| 101 |
-
[
|
| 102 |
-
2304,
|
| 103 |
-
384
|
| 104 |
-
],
|
| 105 |
-
[
|
| 106 |
-
2688,
|
| 107 |
-
384
|
| 108 |
-
],
|
| 109 |
-
[
|
| 110 |
-
3072,
|
| 111 |
-
384
|
| 112 |
-
],
|
| 113 |
-
[
|
| 114 |
-
3456,
|
| 115 |
-
384
|
| 116 |
-
],
|
| 117 |
-
[
|
| 118 |
-
3840,
|
| 119 |
-
384
|
| 120 |
-
]
|
| 121 |
-
],
|
| 122 |
-
"image_seq_length": 576,
|
| 123 |
-
"image_token_index": 49155,
|
| 124 |
-
"model_type": "granitevisionemb",
|
| 125 |
-
"multimodal_projector_bias": true,
|
| 126 |
-
"pretrained_language_model": "",
|
| 127 |
-
"pretrained_vision_tower": "",
|
| 128 |
-
"projector_hidden_act": "gelu",
|
| 129 |
-
"text_config": {
|
| 130 |
-
"_attn_implementation_autoset": true,
|
| 131 |
-
"_name_or_path": "ibm-granite/granite-3.1-2b-instruct",
|
| 132 |
-
"architectures": [
|
| 133 |
-
"GraniteForCausalLM"
|
| 134 |
-
],
|
| 135 |
-
"attention_dropout": 0.1,
|
| 136 |
-
"attention_multiplier": 0.015625,
|
| 137 |
-
"bos_token_id": 0,
|
| 138 |
-
"embedding_multiplier": 12.0,
|
| 139 |
-
"eos_token_id": 0,
|
| 140 |
-
"hidden_size": 2048,
|
| 141 |
-
"intermediate_size": 8192,
|
| 142 |
-
"logits_scaling": 8.0,
|
| 143 |
-
"max_position_embeddings": 131072,
|
| 144 |
-
"model_type": "granite",
|
| 145 |
-
"num_hidden_layers": 40,
|
| 146 |
-
"num_key_value_heads": 8,
|
| 147 |
-
"pad_token_id": 0,
|
| 148 |
-
"residual_multiplier": 0.22,
|
| 149 |
-
"rms_norm_eps": 1e-05,
|
| 150 |
-
"rope_theta": 300000,
|
| 151 |
-
"tie_word_embeddings": true,
|
| 152 |
-
"torch_dtype": "bfloat16",
|
| 153 |
-
"vocab_size": 49156
|
| 154 |
-
},
|
| 155 |
-
"tie_word_embeddings": true,
|
| 156 |
-
"torch_dtype": "float32",
|
| 157 |
-
"transformers_version": "4.49.0",
|
| 158 |
-
"use_image_newline_parameter": true,
|
| 159 |
-
"vision_config": {
|
| 160 |
-
"_attn_implementation_autoset": true,
|
| 161 |
-
"hidden_act": "gelu_pytorch_tanh",
|
| 162 |
-
"hidden_size": 1152,
|
| 163 |
-
"image_size": 384,
|
| 164 |
-
"intermediate_size": 4304,
|
| 165 |
-
"layer_norm_eps": 1e-06,
|
| 166 |
-
"model_type": "siglip_vision_model",
|
| 167 |
-
"num_attention_heads": 16,
|
| 168 |
-
"num_hidden_layers": 27,
|
| 169 |
-
"patch_size": 14,
|
| 170 |
-
"torch_dtype": "bfloat16"
|
| 171 |
-
},
|
| 172 |
-
"vision_feature_layer": [
|
| 173 |
-
-24,
|
| 174 |
-
-20,
|
| 175 |
-
-12,
|
| 176 |
-
-1
|
| 177 |
-
],
|
| 178 |
-
"vision_feature_select_strategy": "full"
|
| 179 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
granite_vision_embedding_config.py
DELETED
|
@@ -1,15 +0,0 @@
|
|
| 1 |
-
from transformers import LlavaNextConfig
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
class GraniteVisionEmbConfig(LlavaNextConfig):
|
| 5 |
-
model_type = "granitevisionemb"
|
| 6 |
-
|
| 7 |
-
def __init__(self, **kwargs):
|
| 8 |
-
self.base_model = kwargs.get("base_model", None)
|
| 9 |
-
self.emb_dim_query = kwargs.get("emb_dim_query", 128)
|
| 10 |
-
self.emb_dim_doc = kwargs.get("emb_dim_doc", 128)
|
| 11 |
-
self.base_image_feature_location = kwargs.get("base_image_feature_location", "last")
|
| 12 |
-
self.adapter_path = kwargs.get("adapter_path", None)
|
| 13 |
-
super().__init__(**kwargs)
|
| 14 |
-
|
| 15 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
merges.txt
DELETED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model-00001-of-00003.safetensors
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:4e838b6d98f48fbf45ae6c0d9c74cba649fd06b27ed78ced3971efbab7e16a69
|
| 3 |
-
size 4955415688
|
|
|
|
|
|
|
|
|
|
|
|
model-00002-of-00003.safetensors
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:d6bf1675fc15977b4d8f37ea1d4960ca2750e6793a80da9771e4693ae8cb13d6
|
| 3 |
-
size 4999979448
|
|
|
|
|
|
|
|
|
|
|
|
model-00003-of-00003.safetensors
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:15978cba0606360676faad5c3cf486a58e6d78a1352dbfcd1db51a7410a574d5
|
| 3 |
-
size 1947355456
|
|
|
|
|
|
|
|
|
|
|
|
model.safetensors.index.json
DELETED
|
@@ -1,824 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"metadata": {
|
| 3 |
-
"total_size": 11902636800
|
| 4 |
-
},
|
| 5 |
-
"weight_map": {
|
| 6 |
-
"custom_text_proj.bias": "model-00003-of-00003.safetensors",
|
| 7 |
-
"custom_text_proj.weight": "model-00003-of-00003.safetensors",
|
| 8 |
-
"model.image_newline": "model-00001-of-00003.safetensors",
|
| 9 |
-
"model.language_model.model.embed_tokens.weight": "model-00001-of-00003.safetensors",
|
| 10 |
-
"model.language_model.model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 11 |
-
"model.language_model.model.layers.0.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 12 |
-
"model.language_model.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 13 |
-
"model.language_model.model.layers.0.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 14 |
-
"model.language_model.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 15 |
-
"model.language_model.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 16 |
-
"model.language_model.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 17 |
-
"model.language_model.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 18 |
-
"model.language_model.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 19 |
-
"model.language_model.model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 20 |
-
"model.language_model.model.layers.1.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 21 |
-
"model.language_model.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 22 |
-
"model.language_model.model.layers.1.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 23 |
-
"model.language_model.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 24 |
-
"model.language_model.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 25 |
-
"model.language_model.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 26 |
-
"model.language_model.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 27 |
-
"model.language_model.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 28 |
-
"model.language_model.model.layers.10.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 29 |
-
"model.language_model.model.layers.10.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 30 |
-
"model.language_model.model.layers.10.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 31 |
-
"model.language_model.model.layers.10.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 32 |
-
"model.language_model.model.layers.10.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 33 |
-
"model.language_model.model.layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 34 |
-
"model.language_model.model.layers.10.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 35 |
-
"model.language_model.model.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 36 |
-
"model.language_model.model.layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 37 |
-
"model.language_model.model.layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 38 |
-
"model.language_model.model.layers.11.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 39 |
-
"model.language_model.model.layers.11.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 40 |
-
"model.language_model.model.layers.11.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 41 |
-
"model.language_model.model.layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 42 |
-
"model.language_model.model.layers.11.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 43 |
-
"model.language_model.model.layers.11.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 44 |
-
"model.language_model.model.layers.11.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 45 |
-
"model.language_model.model.layers.11.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 46 |
-
"model.language_model.model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 47 |
-
"model.language_model.model.layers.12.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 48 |
-
"model.language_model.model.layers.12.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 49 |
-
"model.language_model.model.layers.12.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 50 |
-
"model.language_model.model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 51 |
-
"model.language_model.model.layers.12.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 52 |
-
"model.language_model.model.layers.12.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 53 |
-
"model.language_model.model.layers.12.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 54 |
-
"model.language_model.model.layers.12.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 55 |
-
"model.language_model.model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 56 |
-
"model.language_model.model.layers.13.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 57 |
-
"model.language_model.model.layers.13.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 58 |
-
"model.language_model.model.layers.13.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 59 |
-
"model.language_model.model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 60 |
-
"model.language_model.model.layers.13.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 61 |
-
"model.language_model.model.layers.13.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 62 |
-
"model.language_model.model.layers.13.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 63 |
-
"model.language_model.model.layers.13.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 64 |
-
"model.language_model.model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 65 |
-
"model.language_model.model.layers.14.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 66 |
-
"model.language_model.model.layers.14.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 67 |
-
"model.language_model.model.layers.14.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 68 |
-
"model.language_model.model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 69 |
-
"model.language_model.model.layers.14.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 70 |
-
"model.language_model.model.layers.14.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 71 |
-
"model.language_model.model.layers.14.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 72 |
-
"model.language_model.model.layers.14.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 73 |
-
"model.language_model.model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 74 |
-
"model.language_model.model.layers.15.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 75 |
-
"model.language_model.model.layers.15.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 76 |
-
"model.language_model.model.layers.15.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 77 |
-
"model.language_model.model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 78 |
-
"model.language_model.model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 79 |
-
"model.language_model.model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 80 |
-
"model.language_model.model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 81 |
-
"model.language_model.model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 82 |
-
"model.language_model.model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 83 |
-
"model.language_model.model.layers.16.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 84 |
-
"model.language_model.model.layers.16.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 85 |
-
"model.language_model.model.layers.16.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 86 |
-
"model.language_model.model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 87 |
-
"model.language_model.model.layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 88 |
-
"model.language_model.model.layers.16.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 89 |
-
"model.language_model.model.layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 90 |
-
"model.language_model.model.layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 91 |
-
"model.language_model.model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 92 |
-
"model.language_model.model.layers.17.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 93 |
-
"model.language_model.model.layers.17.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 94 |
-
"model.language_model.model.layers.17.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 95 |
-
"model.language_model.model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 96 |
-
"model.language_model.model.layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 97 |
-
"model.language_model.model.layers.17.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 98 |
-
"model.language_model.model.layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 99 |
-
"model.language_model.model.layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 100 |
-
"model.language_model.model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 101 |
-
"model.language_model.model.layers.18.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 102 |
-
"model.language_model.model.layers.18.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 103 |
-
"model.language_model.model.layers.18.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 104 |
-
"model.language_model.model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 105 |
-
"model.language_model.model.layers.18.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 106 |
-
"model.language_model.model.layers.18.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 107 |
-
"model.language_model.model.layers.18.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 108 |
-
"model.language_model.model.layers.18.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 109 |
-
"model.language_model.model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 110 |
-
"model.language_model.model.layers.19.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 111 |
-
"model.language_model.model.layers.19.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 112 |
-
"model.language_model.model.layers.19.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 113 |
-
"model.language_model.model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 114 |
-
"model.language_model.model.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 115 |
-
"model.language_model.model.layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 116 |
-
"model.language_model.model.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 117 |
-
"model.language_model.model.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 118 |
-
"model.language_model.model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 119 |
-
"model.language_model.model.layers.2.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 120 |
-
"model.language_model.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 121 |
-
"model.language_model.model.layers.2.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 122 |
-
"model.language_model.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 123 |
-
"model.language_model.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 124 |
-
"model.language_model.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 125 |
-
"model.language_model.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 126 |
-
"model.language_model.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 127 |
-
"model.language_model.model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 128 |
-
"model.language_model.model.layers.20.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 129 |
-
"model.language_model.model.layers.20.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 130 |
-
"model.language_model.model.layers.20.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 131 |
-
"model.language_model.model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 132 |
-
"model.language_model.model.layers.20.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 133 |
-
"model.language_model.model.layers.20.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 134 |
-
"model.language_model.model.layers.20.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 135 |
-
"model.language_model.model.layers.20.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 136 |
-
"model.language_model.model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 137 |
-
"model.language_model.model.layers.21.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 138 |
-
"model.language_model.model.layers.21.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 139 |
-
"model.language_model.model.layers.21.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 140 |
-
"model.language_model.model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 141 |
-
"model.language_model.model.layers.21.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 142 |
-
"model.language_model.model.layers.21.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 143 |
-
"model.language_model.model.layers.21.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 144 |
-
"model.language_model.model.layers.21.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 145 |
-
"model.language_model.model.layers.22.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 146 |
-
"model.language_model.model.layers.22.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 147 |
-
"model.language_model.model.layers.22.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 148 |
-
"model.language_model.model.layers.22.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 149 |
-
"model.language_model.model.layers.22.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 150 |
-
"model.language_model.model.layers.22.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 151 |
-
"model.language_model.model.layers.22.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 152 |
-
"model.language_model.model.layers.22.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 153 |
-
"model.language_model.model.layers.22.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 154 |
-
"model.language_model.model.layers.23.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 155 |
-
"model.language_model.model.layers.23.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 156 |
-
"model.language_model.model.layers.23.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 157 |
-
"model.language_model.model.layers.23.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 158 |
-
"model.language_model.model.layers.23.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 159 |
-
"model.language_model.model.layers.23.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 160 |
-
"model.language_model.model.layers.23.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 161 |
-
"model.language_model.model.layers.23.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 162 |
-
"model.language_model.model.layers.23.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 163 |
-
"model.language_model.model.layers.24.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 164 |
-
"model.language_model.model.layers.24.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 165 |
-
"model.language_model.model.layers.24.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 166 |
-
"model.language_model.model.layers.24.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 167 |
-
"model.language_model.model.layers.24.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 168 |
-
"model.language_model.model.layers.24.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 169 |
-
"model.language_model.model.layers.24.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 170 |
-
"model.language_model.model.layers.24.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 171 |
-
"model.language_model.model.layers.24.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 172 |
-
"model.language_model.model.layers.25.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 173 |
-
"model.language_model.model.layers.25.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 174 |
-
"model.language_model.model.layers.25.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 175 |
-
"model.language_model.model.layers.25.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 176 |
-
"model.language_model.model.layers.25.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 177 |
-
"model.language_model.model.layers.25.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 178 |
-
"model.language_model.model.layers.25.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 179 |
-
"model.language_model.model.layers.25.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 180 |
-
"model.language_model.model.layers.25.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 181 |
-
"model.language_model.model.layers.26.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 182 |
-
"model.language_model.model.layers.26.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 183 |
-
"model.language_model.model.layers.26.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 184 |
-
"model.language_model.model.layers.26.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 185 |
-
"model.language_model.model.layers.26.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 186 |
-
"model.language_model.model.layers.26.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 187 |
-
"model.language_model.model.layers.26.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 188 |
-
"model.language_model.model.layers.26.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 189 |
-
"model.language_model.model.layers.26.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 190 |
-
"model.language_model.model.layers.27.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 191 |
-
"model.language_model.model.layers.27.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 192 |
-
"model.language_model.model.layers.27.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 193 |
-
"model.language_model.model.layers.27.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 194 |
-
"model.language_model.model.layers.27.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 195 |
-
"model.language_model.model.layers.27.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 196 |
-
"model.language_model.model.layers.27.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 197 |
-
"model.language_model.model.layers.27.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 198 |
-
"model.language_model.model.layers.27.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 199 |
-
"model.language_model.model.layers.28.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 200 |
-
"model.language_model.model.layers.28.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 201 |
-
"model.language_model.model.layers.28.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 202 |
-
"model.language_model.model.layers.28.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 203 |
-
"model.language_model.model.layers.28.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 204 |
-
"model.language_model.model.layers.28.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 205 |
-
"model.language_model.model.layers.28.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 206 |
-
"model.language_model.model.layers.28.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 207 |
-
"model.language_model.model.layers.28.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 208 |
-
"model.language_model.model.layers.29.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 209 |
-
"model.language_model.model.layers.29.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 210 |
-
"model.language_model.model.layers.29.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 211 |
-
"model.language_model.model.layers.29.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 212 |
-
"model.language_model.model.layers.29.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 213 |
-
"model.language_model.model.layers.29.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 214 |
-
"model.language_model.model.layers.29.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 215 |
-
"model.language_model.model.layers.29.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 216 |
-
"model.language_model.model.layers.29.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 217 |
-
"model.language_model.model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 218 |
-
"model.language_model.model.layers.3.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 219 |
-
"model.language_model.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 220 |
-
"model.language_model.model.layers.3.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 221 |
-
"model.language_model.model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 222 |
-
"model.language_model.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 223 |
-
"model.language_model.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 224 |
-
"model.language_model.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 225 |
-
"model.language_model.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 226 |
-
"model.language_model.model.layers.30.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 227 |
-
"model.language_model.model.layers.30.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 228 |
-
"model.language_model.model.layers.30.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 229 |
-
"model.language_model.model.layers.30.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 230 |
-
"model.language_model.model.layers.30.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 231 |
-
"model.language_model.model.layers.30.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 232 |
-
"model.language_model.model.layers.30.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 233 |
-
"model.language_model.model.layers.30.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 234 |
-
"model.language_model.model.layers.30.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 235 |
-
"model.language_model.model.layers.31.input_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 236 |
-
"model.language_model.model.layers.31.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
|
| 237 |
-
"model.language_model.model.layers.31.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
|
| 238 |
-
"model.language_model.model.layers.31.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
|
| 239 |
-
"model.language_model.model.layers.31.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
|
| 240 |
-
"model.language_model.model.layers.31.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
|
| 241 |
-
"model.language_model.model.layers.31.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
|
| 242 |
-
"model.language_model.model.layers.31.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
|
| 243 |
-
"model.language_model.model.layers.31.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
|
| 244 |
-
"model.language_model.model.layers.32.input_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 245 |
-
"model.language_model.model.layers.32.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
|
| 246 |
-
"model.language_model.model.layers.32.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
|
| 247 |
-
"model.language_model.model.layers.32.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
|
| 248 |
-
"model.language_model.model.layers.32.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 249 |
-
"model.language_model.model.layers.32.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
|
| 250 |
-
"model.language_model.model.layers.32.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
|
| 251 |
-
"model.language_model.model.layers.32.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
|
| 252 |
-
"model.language_model.model.layers.32.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
|
| 253 |
-
"model.language_model.model.layers.33.input_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 254 |
-
"model.language_model.model.layers.33.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
|
| 255 |
-
"model.language_model.model.layers.33.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
|
| 256 |
-
"model.language_model.model.layers.33.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
|
| 257 |
-
"model.language_model.model.layers.33.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 258 |
-
"model.language_model.model.layers.33.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
|
| 259 |
-
"model.language_model.model.layers.33.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
|
| 260 |
-
"model.language_model.model.layers.33.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
|
| 261 |
-
"model.language_model.model.layers.33.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
|
| 262 |
-
"model.language_model.model.layers.34.input_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 263 |
-
"model.language_model.model.layers.34.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
|
| 264 |
-
"model.language_model.model.layers.34.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
|
| 265 |
-
"model.language_model.model.layers.34.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
|
| 266 |
-
"model.language_model.model.layers.34.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 267 |
-
"model.language_model.model.layers.34.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
|
| 268 |
-
"model.language_model.model.layers.34.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
|
| 269 |
-
"model.language_model.model.layers.34.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
|
| 270 |
-
"model.language_model.model.layers.34.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
|
| 271 |
-
"model.language_model.model.layers.35.input_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 272 |
-
"model.language_model.model.layers.35.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
|
| 273 |
-
"model.language_model.model.layers.35.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
|
| 274 |
-
"model.language_model.model.layers.35.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
|
| 275 |
-
"model.language_model.model.layers.35.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 276 |
-
"model.language_model.model.layers.35.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
|
| 277 |
-
"model.language_model.model.layers.35.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
|
| 278 |
-
"model.language_model.model.layers.35.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
|
| 279 |
-
"model.language_model.model.layers.35.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
|
| 280 |
-
"model.language_model.model.layers.36.input_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 281 |
-
"model.language_model.model.layers.36.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
|
| 282 |
-
"model.language_model.model.layers.36.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
|
| 283 |
-
"model.language_model.model.layers.36.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
|
| 284 |
-
"model.language_model.model.layers.36.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 285 |
-
"model.language_model.model.layers.36.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
|
| 286 |
-
"model.language_model.model.layers.36.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
|
| 287 |
-
"model.language_model.model.layers.36.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
|
| 288 |
-
"model.language_model.model.layers.36.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
|
| 289 |
-
"model.language_model.model.layers.37.input_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 290 |
-
"model.language_model.model.layers.37.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
|
| 291 |
-
"model.language_model.model.layers.37.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
|
| 292 |
-
"model.language_model.model.layers.37.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
|
| 293 |
-
"model.language_model.model.layers.37.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 294 |
-
"model.language_model.model.layers.37.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
|
| 295 |
-
"model.language_model.model.layers.37.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
|
| 296 |
-
"model.language_model.model.layers.37.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
|
| 297 |
-
"model.language_model.model.layers.37.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
|
| 298 |
-
"model.language_model.model.layers.38.input_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 299 |
-
"model.language_model.model.layers.38.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
|
| 300 |
-
"model.language_model.model.layers.38.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
|
| 301 |
-
"model.language_model.model.layers.38.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
|
| 302 |
-
"model.language_model.model.layers.38.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 303 |
-
"model.language_model.model.layers.38.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
|
| 304 |
-
"model.language_model.model.layers.38.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
|
| 305 |
-
"model.language_model.model.layers.38.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
|
| 306 |
-
"model.language_model.model.layers.38.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
|
| 307 |
-
"model.language_model.model.layers.39.input_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 308 |
-
"model.language_model.model.layers.39.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
|
| 309 |
-
"model.language_model.model.layers.39.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
|
| 310 |
-
"model.language_model.model.layers.39.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
|
| 311 |
-
"model.language_model.model.layers.39.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
|
| 312 |
-
"model.language_model.model.layers.39.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
|
| 313 |
-
"model.language_model.model.layers.39.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
|
| 314 |
-
"model.language_model.model.layers.39.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
|
| 315 |
-
"model.language_model.model.layers.39.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
|
| 316 |
-
"model.language_model.model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 317 |
-
"model.language_model.model.layers.4.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 318 |
-
"model.language_model.model.layers.4.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 319 |
-
"model.language_model.model.layers.4.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 320 |
-
"model.language_model.model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 321 |
-
"model.language_model.model.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 322 |
-
"model.language_model.model.layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 323 |
-
"model.language_model.model.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 324 |
-
"model.language_model.model.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 325 |
-
"model.language_model.model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 326 |
-
"model.language_model.model.layers.5.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 327 |
-
"model.language_model.model.layers.5.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 328 |
-
"model.language_model.model.layers.5.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 329 |
-
"model.language_model.model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 330 |
-
"model.language_model.model.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 331 |
-
"model.language_model.model.layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 332 |
-
"model.language_model.model.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 333 |
-
"model.language_model.model.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 334 |
-
"model.language_model.model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 335 |
-
"model.language_model.model.layers.6.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 336 |
-
"model.language_model.model.layers.6.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 337 |
-
"model.language_model.model.layers.6.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 338 |
-
"model.language_model.model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 339 |
-
"model.language_model.model.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 340 |
-
"model.language_model.model.layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 341 |
-
"model.language_model.model.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 342 |
-
"model.language_model.model.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 343 |
-
"model.language_model.model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 344 |
-
"model.language_model.model.layers.7.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 345 |
-
"model.language_model.model.layers.7.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 346 |
-
"model.language_model.model.layers.7.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 347 |
-
"model.language_model.model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 348 |
-
"model.language_model.model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 349 |
-
"model.language_model.model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 350 |
-
"model.language_model.model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 351 |
-
"model.language_model.model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 352 |
-
"model.language_model.model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 353 |
-
"model.language_model.model.layers.8.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 354 |
-
"model.language_model.model.layers.8.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 355 |
-
"model.language_model.model.layers.8.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 356 |
-
"model.language_model.model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 357 |
-
"model.language_model.model.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 358 |
-
"model.language_model.model.layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 359 |
-
"model.language_model.model.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 360 |
-
"model.language_model.model.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 361 |
-
"model.language_model.model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 362 |
-
"model.language_model.model.layers.9.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
|
| 363 |
-
"model.language_model.model.layers.9.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
|
| 364 |
-
"model.language_model.model.layers.9.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
|
| 365 |
-
"model.language_model.model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
|
| 366 |
-
"model.language_model.model.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 367 |
-
"model.language_model.model.layers.9.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
|
| 368 |
-
"model.language_model.model.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 369 |
-
"model.language_model.model.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 370 |
-
"model.language_model.model.norm.weight": "model-00003-of-00003.safetensors",
|
| 371 |
-
"model.multi_modal_projector.linear_1.bias": "model-00001-of-00003.safetensors",
|
| 372 |
-
"model.multi_modal_projector.linear_1.weight": "model-00001-of-00003.safetensors",
|
| 373 |
-
"model.multi_modal_projector.linear_2.bias": "model-00001-of-00003.safetensors",
|
| 374 |
-
"model.multi_modal_projector.linear_2.weight": "model-00001-of-00003.safetensors",
|
| 375 |
-
"model.vision_tower.vision_model.embeddings.patch_embedding.bias": "model-00001-of-00003.safetensors",
|
| 376 |
-
"model.vision_tower.vision_model.embeddings.patch_embedding.weight": "model-00001-of-00003.safetensors",
|
| 377 |
-
"model.vision_tower.vision_model.embeddings.position_embedding.weight": "model-00001-of-00003.safetensors",
|
| 378 |
-
"model.vision_tower.vision_model.encoder.layers.0.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 379 |
-
"model.vision_tower.vision_model.encoder.layers.0.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 380 |
-
"model.vision_tower.vision_model.encoder.layers.0.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 381 |
-
"model.vision_tower.vision_model.encoder.layers.0.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 382 |
-
"model.vision_tower.vision_model.encoder.layers.0.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 383 |
-
"model.vision_tower.vision_model.encoder.layers.0.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 384 |
-
"model.vision_tower.vision_model.encoder.layers.0.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 385 |
-
"model.vision_tower.vision_model.encoder.layers.0.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 386 |
-
"model.vision_tower.vision_model.encoder.layers.0.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 387 |
-
"model.vision_tower.vision_model.encoder.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 388 |
-
"model.vision_tower.vision_model.encoder.layers.0.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 389 |
-
"model.vision_tower.vision_model.encoder.layers.0.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 390 |
-
"model.vision_tower.vision_model.encoder.layers.0.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 391 |
-
"model.vision_tower.vision_model.encoder.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 392 |
-
"model.vision_tower.vision_model.encoder.layers.0.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 393 |
-
"model.vision_tower.vision_model.encoder.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 394 |
-
"model.vision_tower.vision_model.encoder.layers.1.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 395 |
-
"model.vision_tower.vision_model.encoder.layers.1.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 396 |
-
"model.vision_tower.vision_model.encoder.layers.1.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 397 |
-
"model.vision_tower.vision_model.encoder.layers.1.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 398 |
-
"model.vision_tower.vision_model.encoder.layers.1.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 399 |
-
"model.vision_tower.vision_model.encoder.layers.1.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 400 |
-
"model.vision_tower.vision_model.encoder.layers.1.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 401 |
-
"model.vision_tower.vision_model.encoder.layers.1.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 402 |
-
"model.vision_tower.vision_model.encoder.layers.1.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 403 |
-
"model.vision_tower.vision_model.encoder.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 404 |
-
"model.vision_tower.vision_model.encoder.layers.1.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 405 |
-
"model.vision_tower.vision_model.encoder.layers.1.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 406 |
-
"model.vision_tower.vision_model.encoder.layers.1.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 407 |
-
"model.vision_tower.vision_model.encoder.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 408 |
-
"model.vision_tower.vision_model.encoder.layers.1.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 409 |
-
"model.vision_tower.vision_model.encoder.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 410 |
-
"model.vision_tower.vision_model.encoder.layers.10.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 411 |
-
"model.vision_tower.vision_model.encoder.layers.10.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 412 |
-
"model.vision_tower.vision_model.encoder.layers.10.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 413 |
-
"model.vision_tower.vision_model.encoder.layers.10.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 414 |
-
"model.vision_tower.vision_model.encoder.layers.10.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 415 |
-
"model.vision_tower.vision_model.encoder.layers.10.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 416 |
-
"model.vision_tower.vision_model.encoder.layers.10.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 417 |
-
"model.vision_tower.vision_model.encoder.layers.10.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 418 |
-
"model.vision_tower.vision_model.encoder.layers.10.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 419 |
-
"model.vision_tower.vision_model.encoder.layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 420 |
-
"model.vision_tower.vision_model.encoder.layers.10.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 421 |
-
"model.vision_tower.vision_model.encoder.layers.10.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 422 |
-
"model.vision_tower.vision_model.encoder.layers.10.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 423 |
-
"model.vision_tower.vision_model.encoder.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 424 |
-
"model.vision_tower.vision_model.encoder.layers.10.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 425 |
-
"model.vision_tower.vision_model.encoder.layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 426 |
-
"model.vision_tower.vision_model.encoder.layers.11.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 427 |
-
"model.vision_tower.vision_model.encoder.layers.11.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 428 |
-
"model.vision_tower.vision_model.encoder.layers.11.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 429 |
-
"model.vision_tower.vision_model.encoder.layers.11.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 430 |
-
"model.vision_tower.vision_model.encoder.layers.11.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 431 |
-
"model.vision_tower.vision_model.encoder.layers.11.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 432 |
-
"model.vision_tower.vision_model.encoder.layers.11.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 433 |
-
"model.vision_tower.vision_model.encoder.layers.11.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 434 |
-
"model.vision_tower.vision_model.encoder.layers.11.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 435 |
-
"model.vision_tower.vision_model.encoder.layers.11.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 436 |
-
"model.vision_tower.vision_model.encoder.layers.11.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 437 |
-
"model.vision_tower.vision_model.encoder.layers.11.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 438 |
-
"model.vision_tower.vision_model.encoder.layers.11.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 439 |
-
"model.vision_tower.vision_model.encoder.layers.11.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 440 |
-
"model.vision_tower.vision_model.encoder.layers.11.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 441 |
-
"model.vision_tower.vision_model.encoder.layers.11.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 442 |
-
"model.vision_tower.vision_model.encoder.layers.12.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 443 |
-
"model.vision_tower.vision_model.encoder.layers.12.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 444 |
-
"model.vision_tower.vision_model.encoder.layers.12.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 445 |
-
"model.vision_tower.vision_model.encoder.layers.12.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 446 |
-
"model.vision_tower.vision_model.encoder.layers.12.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 447 |
-
"model.vision_tower.vision_model.encoder.layers.12.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 448 |
-
"model.vision_tower.vision_model.encoder.layers.12.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 449 |
-
"model.vision_tower.vision_model.encoder.layers.12.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 450 |
-
"model.vision_tower.vision_model.encoder.layers.12.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 451 |
-
"model.vision_tower.vision_model.encoder.layers.12.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 452 |
-
"model.vision_tower.vision_model.encoder.layers.12.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 453 |
-
"model.vision_tower.vision_model.encoder.layers.12.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 454 |
-
"model.vision_tower.vision_model.encoder.layers.12.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 455 |
-
"model.vision_tower.vision_model.encoder.layers.12.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 456 |
-
"model.vision_tower.vision_model.encoder.layers.12.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 457 |
-
"model.vision_tower.vision_model.encoder.layers.12.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 458 |
-
"model.vision_tower.vision_model.encoder.layers.13.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 459 |
-
"model.vision_tower.vision_model.encoder.layers.13.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 460 |
-
"model.vision_tower.vision_model.encoder.layers.13.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 461 |
-
"model.vision_tower.vision_model.encoder.layers.13.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 462 |
-
"model.vision_tower.vision_model.encoder.layers.13.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 463 |
-
"model.vision_tower.vision_model.encoder.layers.13.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 464 |
-
"model.vision_tower.vision_model.encoder.layers.13.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 465 |
-
"model.vision_tower.vision_model.encoder.layers.13.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 466 |
-
"model.vision_tower.vision_model.encoder.layers.13.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 467 |
-
"model.vision_tower.vision_model.encoder.layers.13.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 468 |
-
"model.vision_tower.vision_model.encoder.layers.13.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 469 |
-
"model.vision_tower.vision_model.encoder.layers.13.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 470 |
-
"model.vision_tower.vision_model.encoder.layers.13.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 471 |
-
"model.vision_tower.vision_model.encoder.layers.13.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 472 |
-
"model.vision_tower.vision_model.encoder.layers.13.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 473 |
-
"model.vision_tower.vision_model.encoder.layers.13.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 474 |
-
"model.vision_tower.vision_model.encoder.layers.14.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 475 |
-
"model.vision_tower.vision_model.encoder.layers.14.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 476 |
-
"model.vision_tower.vision_model.encoder.layers.14.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 477 |
-
"model.vision_tower.vision_model.encoder.layers.14.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 478 |
-
"model.vision_tower.vision_model.encoder.layers.14.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 479 |
-
"model.vision_tower.vision_model.encoder.layers.14.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 480 |
-
"model.vision_tower.vision_model.encoder.layers.14.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 481 |
-
"model.vision_tower.vision_model.encoder.layers.14.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 482 |
-
"model.vision_tower.vision_model.encoder.layers.14.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 483 |
-
"model.vision_tower.vision_model.encoder.layers.14.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 484 |
-
"model.vision_tower.vision_model.encoder.layers.14.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 485 |
-
"model.vision_tower.vision_model.encoder.layers.14.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 486 |
-
"model.vision_tower.vision_model.encoder.layers.14.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 487 |
-
"model.vision_tower.vision_model.encoder.layers.14.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 488 |
-
"model.vision_tower.vision_model.encoder.layers.14.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 489 |
-
"model.vision_tower.vision_model.encoder.layers.14.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 490 |
-
"model.vision_tower.vision_model.encoder.layers.15.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 491 |
-
"model.vision_tower.vision_model.encoder.layers.15.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 492 |
-
"model.vision_tower.vision_model.encoder.layers.15.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 493 |
-
"model.vision_tower.vision_model.encoder.layers.15.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 494 |
-
"model.vision_tower.vision_model.encoder.layers.15.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 495 |
-
"model.vision_tower.vision_model.encoder.layers.15.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 496 |
-
"model.vision_tower.vision_model.encoder.layers.15.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 497 |
-
"model.vision_tower.vision_model.encoder.layers.15.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 498 |
-
"model.vision_tower.vision_model.encoder.layers.15.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 499 |
-
"model.vision_tower.vision_model.encoder.layers.15.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 500 |
-
"model.vision_tower.vision_model.encoder.layers.15.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 501 |
-
"model.vision_tower.vision_model.encoder.layers.15.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 502 |
-
"model.vision_tower.vision_model.encoder.layers.15.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 503 |
-
"model.vision_tower.vision_model.encoder.layers.15.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 504 |
-
"model.vision_tower.vision_model.encoder.layers.15.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 505 |
-
"model.vision_tower.vision_model.encoder.layers.15.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 506 |
-
"model.vision_tower.vision_model.encoder.layers.16.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 507 |
-
"model.vision_tower.vision_model.encoder.layers.16.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 508 |
-
"model.vision_tower.vision_model.encoder.layers.16.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 509 |
-
"model.vision_tower.vision_model.encoder.layers.16.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 510 |
-
"model.vision_tower.vision_model.encoder.layers.16.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 511 |
-
"model.vision_tower.vision_model.encoder.layers.16.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 512 |
-
"model.vision_tower.vision_model.encoder.layers.16.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 513 |
-
"model.vision_tower.vision_model.encoder.layers.16.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 514 |
-
"model.vision_tower.vision_model.encoder.layers.16.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 515 |
-
"model.vision_tower.vision_model.encoder.layers.16.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 516 |
-
"model.vision_tower.vision_model.encoder.layers.16.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 517 |
-
"model.vision_tower.vision_model.encoder.layers.16.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 518 |
-
"model.vision_tower.vision_model.encoder.layers.16.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 519 |
-
"model.vision_tower.vision_model.encoder.layers.16.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 520 |
-
"model.vision_tower.vision_model.encoder.layers.16.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 521 |
-
"model.vision_tower.vision_model.encoder.layers.16.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 522 |
-
"model.vision_tower.vision_model.encoder.layers.17.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 523 |
-
"model.vision_tower.vision_model.encoder.layers.17.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 524 |
-
"model.vision_tower.vision_model.encoder.layers.17.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 525 |
-
"model.vision_tower.vision_model.encoder.layers.17.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 526 |
-
"model.vision_tower.vision_model.encoder.layers.17.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 527 |
-
"model.vision_tower.vision_model.encoder.layers.17.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 528 |
-
"model.vision_tower.vision_model.encoder.layers.17.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 529 |
-
"model.vision_tower.vision_model.encoder.layers.17.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 530 |
-
"model.vision_tower.vision_model.encoder.layers.17.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 531 |
-
"model.vision_tower.vision_model.encoder.layers.17.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 532 |
-
"model.vision_tower.vision_model.encoder.layers.17.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 533 |
-
"model.vision_tower.vision_model.encoder.layers.17.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 534 |
-
"model.vision_tower.vision_model.encoder.layers.17.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 535 |
-
"model.vision_tower.vision_model.encoder.layers.17.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 536 |
-
"model.vision_tower.vision_model.encoder.layers.17.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 537 |
-
"model.vision_tower.vision_model.encoder.layers.17.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 538 |
-
"model.vision_tower.vision_model.encoder.layers.18.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 539 |
-
"model.vision_tower.vision_model.encoder.layers.18.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 540 |
-
"model.vision_tower.vision_model.encoder.layers.18.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 541 |
-
"model.vision_tower.vision_model.encoder.layers.18.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 542 |
-
"model.vision_tower.vision_model.encoder.layers.18.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 543 |
-
"model.vision_tower.vision_model.encoder.layers.18.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 544 |
-
"model.vision_tower.vision_model.encoder.layers.18.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 545 |
-
"model.vision_tower.vision_model.encoder.layers.18.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 546 |
-
"model.vision_tower.vision_model.encoder.layers.18.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 547 |
-
"model.vision_tower.vision_model.encoder.layers.18.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 548 |
-
"model.vision_tower.vision_model.encoder.layers.18.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 549 |
-
"model.vision_tower.vision_model.encoder.layers.18.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 550 |
-
"model.vision_tower.vision_model.encoder.layers.18.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 551 |
-
"model.vision_tower.vision_model.encoder.layers.18.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 552 |
-
"model.vision_tower.vision_model.encoder.layers.18.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 553 |
-
"model.vision_tower.vision_model.encoder.layers.18.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 554 |
-
"model.vision_tower.vision_model.encoder.layers.19.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 555 |
-
"model.vision_tower.vision_model.encoder.layers.19.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 556 |
-
"model.vision_tower.vision_model.encoder.layers.19.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 557 |
-
"model.vision_tower.vision_model.encoder.layers.19.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 558 |
-
"model.vision_tower.vision_model.encoder.layers.19.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 559 |
-
"model.vision_tower.vision_model.encoder.layers.19.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 560 |
-
"model.vision_tower.vision_model.encoder.layers.19.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 561 |
-
"model.vision_tower.vision_model.encoder.layers.19.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 562 |
-
"model.vision_tower.vision_model.encoder.layers.19.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 563 |
-
"model.vision_tower.vision_model.encoder.layers.19.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 564 |
-
"model.vision_tower.vision_model.encoder.layers.19.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 565 |
-
"model.vision_tower.vision_model.encoder.layers.19.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 566 |
-
"model.vision_tower.vision_model.encoder.layers.19.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 567 |
-
"model.vision_tower.vision_model.encoder.layers.19.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 568 |
-
"model.vision_tower.vision_model.encoder.layers.19.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 569 |
-
"model.vision_tower.vision_model.encoder.layers.19.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 570 |
-
"model.vision_tower.vision_model.encoder.layers.2.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 571 |
-
"model.vision_tower.vision_model.encoder.layers.2.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 572 |
-
"model.vision_tower.vision_model.encoder.layers.2.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 573 |
-
"model.vision_tower.vision_model.encoder.layers.2.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 574 |
-
"model.vision_tower.vision_model.encoder.layers.2.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 575 |
-
"model.vision_tower.vision_model.encoder.layers.2.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 576 |
-
"model.vision_tower.vision_model.encoder.layers.2.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 577 |
-
"model.vision_tower.vision_model.encoder.layers.2.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 578 |
-
"model.vision_tower.vision_model.encoder.layers.2.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 579 |
-
"model.vision_tower.vision_model.encoder.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 580 |
-
"model.vision_tower.vision_model.encoder.layers.2.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 581 |
-
"model.vision_tower.vision_model.encoder.layers.2.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 582 |
-
"model.vision_tower.vision_model.encoder.layers.2.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 583 |
-
"model.vision_tower.vision_model.encoder.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 584 |
-
"model.vision_tower.vision_model.encoder.layers.2.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 585 |
-
"model.vision_tower.vision_model.encoder.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 586 |
-
"model.vision_tower.vision_model.encoder.layers.20.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 587 |
-
"model.vision_tower.vision_model.encoder.layers.20.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 588 |
-
"model.vision_tower.vision_model.encoder.layers.20.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 589 |
-
"model.vision_tower.vision_model.encoder.layers.20.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 590 |
-
"model.vision_tower.vision_model.encoder.layers.20.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 591 |
-
"model.vision_tower.vision_model.encoder.layers.20.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 592 |
-
"model.vision_tower.vision_model.encoder.layers.20.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 593 |
-
"model.vision_tower.vision_model.encoder.layers.20.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 594 |
-
"model.vision_tower.vision_model.encoder.layers.20.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 595 |
-
"model.vision_tower.vision_model.encoder.layers.20.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 596 |
-
"model.vision_tower.vision_model.encoder.layers.20.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 597 |
-
"model.vision_tower.vision_model.encoder.layers.20.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 598 |
-
"model.vision_tower.vision_model.encoder.layers.20.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 599 |
-
"model.vision_tower.vision_model.encoder.layers.20.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 600 |
-
"model.vision_tower.vision_model.encoder.layers.20.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 601 |
-
"model.vision_tower.vision_model.encoder.layers.20.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 602 |
-
"model.vision_tower.vision_model.encoder.layers.21.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 603 |
-
"model.vision_tower.vision_model.encoder.layers.21.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 604 |
-
"model.vision_tower.vision_model.encoder.layers.21.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 605 |
-
"model.vision_tower.vision_model.encoder.layers.21.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 606 |
-
"model.vision_tower.vision_model.encoder.layers.21.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 607 |
-
"model.vision_tower.vision_model.encoder.layers.21.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 608 |
-
"model.vision_tower.vision_model.encoder.layers.21.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 609 |
-
"model.vision_tower.vision_model.encoder.layers.21.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 610 |
-
"model.vision_tower.vision_model.encoder.layers.21.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 611 |
-
"model.vision_tower.vision_model.encoder.layers.21.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 612 |
-
"model.vision_tower.vision_model.encoder.layers.21.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 613 |
-
"model.vision_tower.vision_model.encoder.layers.21.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 614 |
-
"model.vision_tower.vision_model.encoder.layers.21.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 615 |
-
"model.vision_tower.vision_model.encoder.layers.21.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 616 |
-
"model.vision_tower.vision_model.encoder.layers.21.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 617 |
-
"model.vision_tower.vision_model.encoder.layers.21.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 618 |
-
"model.vision_tower.vision_model.encoder.layers.22.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 619 |
-
"model.vision_tower.vision_model.encoder.layers.22.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 620 |
-
"model.vision_tower.vision_model.encoder.layers.22.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 621 |
-
"model.vision_tower.vision_model.encoder.layers.22.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 622 |
-
"model.vision_tower.vision_model.encoder.layers.22.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 623 |
-
"model.vision_tower.vision_model.encoder.layers.22.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 624 |
-
"model.vision_tower.vision_model.encoder.layers.22.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 625 |
-
"model.vision_tower.vision_model.encoder.layers.22.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 626 |
-
"model.vision_tower.vision_model.encoder.layers.22.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 627 |
-
"model.vision_tower.vision_model.encoder.layers.22.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 628 |
-
"model.vision_tower.vision_model.encoder.layers.22.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 629 |
-
"model.vision_tower.vision_model.encoder.layers.22.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 630 |
-
"model.vision_tower.vision_model.encoder.layers.22.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 631 |
-
"model.vision_tower.vision_model.encoder.layers.22.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 632 |
-
"model.vision_tower.vision_model.encoder.layers.22.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 633 |
-
"model.vision_tower.vision_model.encoder.layers.22.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 634 |
-
"model.vision_tower.vision_model.encoder.layers.23.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 635 |
-
"model.vision_tower.vision_model.encoder.layers.23.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 636 |
-
"model.vision_tower.vision_model.encoder.layers.23.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 637 |
-
"model.vision_tower.vision_model.encoder.layers.23.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 638 |
-
"model.vision_tower.vision_model.encoder.layers.23.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 639 |
-
"model.vision_tower.vision_model.encoder.layers.23.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 640 |
-
"model.vision_tower.vision_model.encoder.layers.23.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 641 |
-
"model.vision_tower.vision_model.encoder.layers.23.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 642 |
-
"model.vision_tower.vision_model.encoder.layers.23.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 643 |
-
"model.vision_tower.vision_model.encoder.layers.23.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 644 |
-
"model.vision_tower.vision_model.encoder.layers.23.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 645 |
-
"model.vision_tower.vision_model.encoder.layers.23.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 646 |
-
"model.vision_tower.vision_model.encoder.layers.23.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 647 |
-
"model.vision_tower.vision_model.encoder.layers.23.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 648 |
-
"model.vision_tower.vision_model.encoder.layers.23.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 649 |
-
"model.vision_tower.vision_model.encoder.layers.23.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 650 |
-
"model.vision_tower.vision_model.encoder.layers.24.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 651 |
-
"model.vision_tower.vision_model.encoder.layers.24.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 652 |
-
"model.vision_tower.vision_model.encoder.layers.24.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 653 |
-
"model.vision_tower.vision_model.encoder.layers.24.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 654 |
-
"model.vision_tower.vision_model.encoder.layers.24.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 655 |
-
"model.vision_tower.vision_model.encoder.layers.24.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 656 |
-
"model.vision_tower.vision_model.encoder.layers.24.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 657 |
-
"model.vision_tower.vision_model.encoder.layers.24.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 658 |
-
"model.vision_tower.vision_model.encoder.layers.24.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 659 |
-
"model.vision_tower.vision_model.encoder.layers.24.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 660 |
-
"model.vision_tower.vision_model.encoder.layers.24.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 661 |
-
"model.vision_tower.vision_model.encoder.layers.24.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 662 |
-
"model.vision_tower.vision_model.encoder.layers.24.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 663 |
-
"model.vision_tower.vision_model.encoder.layers.24.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 664 |
-
"model.vision_tower.vision_model.encoder.layers.24.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 665 |
-
"model.vision_tower.vision_model.encoder.layers.24.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 666 |
-
"model.vision_tower.vision_model.encoder.layers.25.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 667 |
-
"model.vision_tower.vision_model.encoder.layers.25.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 668 |
-
"model.vision_tower.vision_model.encoder.layers.25.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 669 |
-
"model.vision_tower.vision_model.encoder.layers.25.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 670 |
-
"model.vision_tower.vision_model.encoder.layers.25.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 671 |
-
"model.vision_tower.vision_model.encoder.layers.25.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 672 |
-
"model.vision_tower.vision_model.encoder.layers.25.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 673 |
-
"model.vision_tower.vision_model.encoder.layers.25.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 674 |
-
"model.vision_tower.vision_model.encoder.layers.25.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 675 |
-
"model.vision_tower.vision_model.encoder.layers.25.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 676 |
-
"model.vision_tower.vision_model.encoder.layers.25.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 677 |
-
"model.vision_tower.vision_model.encoder.layers.25.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 678 |
-
"model.vision_tower.vision_model.encoder.layers.25.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 679 |
-
"model.vision_tower.vision_model.encoder.layers.25.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 680 |
-
"model.vision_tower.vision_model.encoder.layers.25.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 681 |
-
"model.vision_tower.vision_model.encoder.layers.25.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 682 |
-
"model.vision_tower.vision_model.encoder.layers.26.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 683 |
-
"model.vision_tower.vision_model.encoder.layers.26.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 684 |
-
"model.vision_tower.vision_model.encoder.layers.26.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 685 |
-
"model.vision_tower.vision_model.encoder.layers.26.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 686 |
-
"model.vision_tower.vision_model.encoder.layers.26.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 687 |
-
"model.vision_tower.vision_model.encoder.layers.26.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 688 |
-
"model.vision_tower.vision_model.encoder.layers.26.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 689 |
-
"model.vision_tower.vision_model.encoder.layers.26.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 690 |
-
"model.vision_tower.vision_model.encoder.layers.26.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 691 |
-
"model.vision_tower.vision_model.encoder.layers.26.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 692 |
-
"model.vision_tower.vision_model.encoder.layers.26.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 693 |
-
"model.vision_tower.vision_model.encoder.layers.26.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 694 |
-
"model.vision_tower.vision_model.encoder.layers.26.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 695 |
-
"model.vision_tower.vision_model.encoder.layers.26.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 696 |
-
"model.vision_tower.vision_model.encoder.layers.26.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 697 |
-
"model.vision_tower.vision_model.encoder.layers.26.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 698 |
-
"model.vision_tower.vision_model.encoder.layers.3.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 699 |
-
"model.vision_tower.vision_model.encoder.layers.3.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 700 |
-
"model.vision_tower.vision_model.encoder.layers.3.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 701 |
-
"model.vision_tower.vision_model.encoder.layers.3.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 702 |
-
"model.vision_tower.vision_model.encoder.layers.3.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 703 |
-
"model.vision_tower.vision_model.encoder.layers.3.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 704 |
-
"model.vision_tower.vision_model.encoder.layers.3.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 705 |
-
"model.vision_tower.vision_model.encoder.layers.3.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 706 |
-
"model.vision_tower.vision_model.encoder.layers.3.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 707 |
-
"model.vision_tower.vision_model.encoder.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 708 |
-
"model.vision_tower.vision_model.encoder.layers.3.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 709 |
-
"model.vision_tower.vision_model.encoder.layers.3.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 710 |
-
"model.vision_tower.vision_model.encoder.layers.3.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 711 |
-
"model.vision_tower.vision_model.encoder.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 712 |
-
"model.vision_tower.vision_model.encoder.layers.3.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 713 |
-
"model.vision_tower.vision_model.encoder.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 714 |
-
"model.vision_tower.vision_model.encoder.layers.4.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 715 |
-
"model.vision_tower.vision_model.encoder.layers.4.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 716 |
-
"model.vision_tower.vision_model.encoder.layers.4.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 717 |
-
"model.vision_tower.vision_model.encoder.layers.4.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 718 |
-
"model.vision_tower.vision_model.encoder.layers.4.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 719 |
-
"model.vision_tower.vision_model.encoder.layers.4.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 720 |
-
"model.vision_tower.vision_model.encoder.layers.4.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 721 |
-
"model.vision_tower.vision_model.encoder.layers.4.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 722 |
-
"model.vision_tower.vision_model.encoder.layers.4.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 723 |
-
"model.vision_tower.vision_model.encoder.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 724 |
-
"model.vision_tower.vision_model.encoder.layers.4.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 725 |
-
"model.vision_tower.vision_model.encoder.layers.4.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 726 |
-
"model.vision_tower.vision_model.encoder.layers.4.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 727 |
-
"model.vision_tower.vision_model.encoder.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 728 |
-
"model.vision_tower.vision_model.encoder.layers.4.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 729 |
-
"model.vision_tower.vision_model.encoder.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 730 |
-
"model.vision_tower.vision_model.encoder.layers.5.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 731 |
-
"model.vision_tower.vision_model.encoder.layers.5.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 732 |
-
"model.vision_tower.vision_model.encoder.layers.5.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 733 |
-
"model.vision_tower.vision_model.encoder.layers.5.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 734 |
-
"model.vision_tower.vision_model.encoder.layers.5.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 735 |
-
"model.vision_tower.vision_model.encoder.layers.5.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 736 |
-
"model.vision_tower.vision_model.encoder.layers.5.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 737 |
-
"model.vision_tower.vision_model.encoder.layers.5.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 738 |
-
"model.vision_tower.vision_model.encoder.layers.5.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 739 |
-
"model.vision_tower.vision_model.encoder.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 740 |
-
"model.vision_tower.vision_model.encoder.layers.5.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 741 |
-
"model.vision_tower.vision_model.encoder.layers.5.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 742 |
-
"model.vision_tower.vision_model.encoder.layers.5.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 743 |
-
"model.vision_tower.vision_model.encoder.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 744 |
-
"model.vision_tower.vision_model.encoder.layers.5.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 745 |
-
"model.vision_tower.vision_model.encoder.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 746 |
-
"model.vision_tower.vision_model.encoder.layers.6.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 747 |
-
"model.vision_tower.vision_model.encoder.layers.6.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 748 |
-
"model.vision_tower.vision_model.encoder.layers.6.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 749 |
-
"model.vision_tower.vision_model.encoder.layers.6.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 750 |
-
"model.vision_tower.vision_model.encoder.layers.6.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 751 |
-
"model.vision_tower.vision_model.encoder.layers.6.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 752 |
-
"model.vision_tower.vision_model.encoder.layers.6.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 753 |
-
"model.vision_tower.vision_model.encoder.layers.6.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 754 |
-
"model.vision_tower.vision_model.encoder.layers.6.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 755 |
-
"model.vision_tower.vision_model.encoder.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 756 |
-
"model.vision_tower.vision_model.encoder.layers.6.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 757 |
-
"model.vision_tower.vision_model.encoder.layers.6.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 758 |
-
"model.vision_tower.vision_model.encoder.layers.6.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 759 |
-
"model.vision_tower.vision_model.encoder.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 760 |
-
"model.vision_tower.vision_model.encoder.layers.6.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 761 |
-
"model.vision_tower.vision_model.encoder.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 762 |
-
"model.vision_tower.vision_model.encoder.layers.7.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 763 |
-
"model.vision_tower.vision_model.encoder.layers.7.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 764 |
-
"model.vision_tower.vision_model.encoder.layers.7.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 765 |
-
"model.vision_tower.vision_model.encoder.layers.7.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 766 |
-
"model.vision_tower.vision_model.encoder.layers.7.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 767 |
-
"model.vision_tower.vision_model.encoder.layers.7.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 768 |
-
"model.vision_tower.vision_model.encoder.layers.7.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 769 |
-
"model.vision_tower.vision_model.encoder.layers.7.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 770 |
-
"model.vision_tower.vision_model.encoder.layers.7.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 771 |
-
"model.vision_tower.vision_model.encoder.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 772 |
-
"model.vision_tower.vision_model.encoder.layers.7.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 773 |
-
"model.vision_tower.vision_model.encoder.layers.7.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 774 |
-
"model.vision_tower.vision_model.encoder.layers.7.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 775 |
-
"model.vision_tower.vision_model.encoder.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 776 |
-
"model.vision_tower.vision_model.encoder.layers.7.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 777 |
-
"model.vision_tower.vision_model.encoder.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 778 |
-
"model.vision_tower.vision_model.encoder.layers.8.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 779 |
-
"model.vision_tower.vision_model.encoder.layers.8.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 780 |
-
"model.vision_tower.vision_model.encoder.layers.8.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 781 |
-
"model.vision_tower.vision_model.encoder.layers.8.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 782 |
-
"model.vision_tower.vision_model.encoder.layers.8.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 783 |
-
"model.vision_tower.vision_model.encoder.layers.8.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 784 |
-
"model.vision_tower.vision_model.encoder.layers.8.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 785 |
-
"model.vision_tower.vision_model.encoder.layers.8.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 786 |
-
"model.vision_tower.vision_model.encoder.layers.8.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 787 |
-
"model.vision_tower.vision_model.encoder.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 788 |
-
"model.vision_tower.vision_model.encoder.layers.8.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 789 |
-
"model.vision_tower.vision_model.encoder.layers.8.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 790 |
-
"model.vision_tower.vision_model.encoder.layers.8.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 791 |
-
"model.vision_tower.vision_model.encoder.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 792 |
-
"model.vision_tower.vision_model.encoder.layers.8.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 793 |
-
"model.vision_tower.vision_model.encoder.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 794 |
-
"model.vision_tower.vision_model.encoder.layers.9.layer_norm1.bias": "model-00001-of-00003.safetensors",
|
| 795 |
-
"model.vision_tower.vision_model.encoder.layers.9.layer_norm1.weight": "model-00001-of-00003.safetensors",
|
| 796 |
-
"model.vision_tower.vision_model.encoder.layers.9.layer_norm2.bias": "model-00001-of-00003.safetensors",
|
| 797 |
-
"model.vision_tower.vision_model.encoder.layers.9.layer_norm2.weight": "model-00001-of-00003.safetensors",
|
| 798 |
-
"model.vision_tower.vision_model.encoder.layers.9.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 799 |
-
"model.vision_tower.vision_model.encoder.layers.9.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 800 |
-
"model.vision_tower.vision_model.encoder.layers.9.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 801 |
-
"model.vision_tower.vision_model.encoder.layers.9.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 802 |
-
"model.vision_tower.vision_model.encoder.layers.9.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
|
| 803 |
-
"model.vision_tower.vision_model.encoder.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
|
| 804 |
-
"model.vision_tower.vision_model.encoder.layers.9.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 805 |
-
"model.vision_tower.vision_model.encoder.layers.9.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 806 |
-
"model.vision_tower.vision_model.encoder.layers.9.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
|
| 807 |
-
"model.vision_tower.vision_model.encoder.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
|
| 808 |
-
"model.vision_tower.vision_model.encoder.layers.9.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
|
| 809 |
-
"model.vision_tower.vision_model.encoder.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
|
| 810 |
-
"model.vision_tower.vision_model.head.attention.in_proj_bias": "model-00001-of-00003.safetensors",
|
| 811 |
-
"model.vision_tower.vision_model.head.attention.in_proj_weight": "model-00001-of-00003.safetensors",
|
| 812 |
-
"model.vision_tower.vision_model.head.attention.out_proj.bias": "model-00001-of-00003.safetensors",
|
| 813 |
-
"model.vision_tower.vision_model.head.attention.out_proj.weight": "model-00001-of-00003.safetensors",
|
| 814 |
-
"model.vision_tower.vision_model.head.layernorm.bias": "model-00001-of-00003.safetensors",
|
| 815 |
-
"model.vision_tower.vision_model.head.layernorm.weight": "model-00001-of-00003.safetensors",
|
| 816 |
-
"model.vision_tower.vision_model.head.mlp.fc1.bias": "model-00001-of-00003.safetensors",
|
| 817 |
-
"model.vision_tower.vision_model.head.mlp.fc1.weight": "model-00001-of-00003.safetensors",
|
| 818 |
-
"model.vision_tower.vision_model.head.mlp.fc2.bias": "model-00001-of-00003.safetensors",
|
| 819 |
-
"model.vision_tower.vision_model.head.mlp.fc2.weight": "model-00001-of-00003.safetensors",
|
| 820 |
-
"model.vision_tower.vision_model.head.probe": "model-00001-of-00003.safetensors",
|
| 821 |
-
"model.vision_tower.vision_model.post_layernorm.bias": "model-00001-of-00003.safetensors",
|
| 822 |
-
"model.vision_tower.vision_model.post_layernorm.weight": "model-00001-of-00003.safetensors"
|
| 823 |
-
}
|
| 824 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
modeling_granite_vision_embedding.py
DELETED
|
@@ -1,190 +0,0 @@
|
|
| 1 |
-
from typing import ClassVar, Optional
|
| 2 |
-
|
| 3 |
-
import numpy as np
|
| 4 |
-
import torch
|
| 5 |
-
from torch import nn
|
| 6 |
-
from transformers import LlavaNextPreTrainedModel
|
| 7 |
-
from transformers.models.llava_next.modeling_llava_next import LlavaNextForConditionalGeneration
|
| 8 |
-
from transformers.models.llava_next.modeling_llava_next import unpad_image, get_anyres_image_grid_shape
|
| 9 |
-
|
| 10 |
-
from .granite_vision_embedding_config import GraniteVisionEmbConfig
|
| 11 |
-
|
| 12 |
-
class LlavaNextWithCustomPacking(LlavaNextForConditionalGeneration):
|
| 13 |
-
|
| 14 |
-
def pack_image_features(
|
| 15 |
-
self,
|
| 16 |
-
image_features,
|
| 17 |
-
image_sizes,
|
| 18 |
-
vision_feature_select_strategy,
|
| 19 |
-
image_newline=None
|
| 20 |
-
):
|
| 21 |
-
"""
|
| 22 |
-
Reshape, unpad and then pack each image_feature into a single image_features tensor containing all visual vectors.
|
| 23 |
-
|
| 24 |
-
Args:
|
| 25 |
-
image_features (`List[torch.Tensor]` of length num_images, each of shape `(num_patches, image_length, embed_dim)`)
|
| 26 |
-
List of image feature tensor, each contains all the visual feature of all patches.
|
| 27 |
-
image_sizes (`torch.Tensor` of shape `(num_images, 2)`)
|
| 28 |
-
Actual image size of each images (H, W).
|
| 29 |
-
vision_feature_select_strategy (`str`)
|
| 30 |
-
The feature selection strategy used to select the vision feature from the vision backbone.
|
| 31 |
-
image_newline (`torch.Tensor` of shape `(embed_dim)`)
|
| 32 |
-
New line embedding vector.
|
| 33 |
-
Returns:
|
| 34 |
-
image_features (`torch.Tensor` of shape `(all_feat_len, embed_dim)`)
|
| 35 |
-
feature_lens (`List[int]`)
|
| 36 |
-
token length of each image in image_features
|
| 37 |
-
"""
|
| 38 |
-
|
| 39 |
-
base_image_feature_location = self.config.base_image_feature_location
|
| 40 |
-
new_image_features = []
|
| 41 |
-
feature_lens = []
|
| 42 |
-
for image_idx, image_feature in enumerate(image_features):
|
| 43 |
-
if image_feature.shape[0] > 1:
|
| 44 |
-
base_image_feature = image_feature[0]
|
| 45 |
-
image_feature = image_feature[1:]
|
| 46 |
-
height = width = self.config.vision_config.image_size // self.config.vision_config.patch_size
|
| 47 |
-
|
| 48 |
-
num_patch_height, num_patch_width = get_anyres_image_grid_shape(
|
| 49 |
-
image_sizes[image_idx],
|
| 50 |
-
self.config.image_grid_pinpoints,
|
| 51 |
-
self.config.vision_config.image_size,
|
| 52 |
-
)
|
| 53 |
-
|
| 54 |
-
if (
|
| 55 |
-
np.prod(image_feature.shape) % (num_patch_height * num_patch_width * height * width) != 0
|
| 56 |
-
and vision_feature_select_strategy == "default"
|
| 57 |
-
):
|
| 58 |
-
print(
|
| 59 |
-
"Image feature shape does not line up with the provided patch size. "
|
| 60 |
-
"You may be using the `default` vision_feature_select_strategy with a"
|
| 61 |
-
" visual encoder that does not have CLS."
|
| 62 |
-
)
|
| 63 |
-
|
| 64 |
-
image_feature = image_feature.view(num_patch_height, num_patch_width, height, width, -1)
|
| 65 |
-
image_feature = image_feature.permute(4, 0, 2, 1, 3).contiguous()
|
| 66 |
-
image_feature = image_feature.flatten(1, 2).flatten(2, 3)
|
| 67 |
-
image_feature = unpad_image(image_feature, image_sizes[image_idx])
|
| 68 |
-
if image_newline is not None:
|
| 69 |
-
image_feature = torch.cat(
|
| 70 |
-
(
|
| 71 |
-
image_feature,
|
| 72 |
-
image_newline[:, None, None]
|
| 73 |
-
.expand(*image_feature.shape[:-1], 1)
|
| 74 |
-
.to(image_feature.device, image_feature.dtype),
|
| 75 |
-
),
|
| 76 |
-
dim=-1,
|
| 77 |
-
)
|
| 78 |
-
image_feature = image_feature.flatten(1, 2).transpose(0, 1)
|
| 79 |
-
if base_image_feature_location == "last":
|
| 80 |
-
image_feature = torch.cat((image_feature, base_image_feature), dim=0)
|
| 81 |
-
else:
|
| 82 |
-
image_feature = torch.cat((base_image_feature, image_feature), dim=0)
|
| 83 |
-
|
| 84 |
-
else:
|
| 85 |
-
image_feature = image_feature[0]
|
| 86 |
-
if image_newline is not None:
|
| 87 |
-
image_feature = torch.cat((image_feature, image_newline[None].to(image_feature)), dim=0)
|
| 88 |
-
new_image_features.append(image_feature)
|
| 89 |
-
feature_lens.append(image_feature.size(0))
|
| 90 |
-
image_features = torch.cat(new_image_features, dim=0)
|
| 91 |
-
feature_lens = torch.tensor(feature_lens, dtype=torch.long, device=image_features.device)
|
| 92 |
-
return image_features, feature_lens
|
| 93 |
-
|
| 94 |
-
|
| 95 |
-
class GraniteVisionEmb(LlavaNextPreTrainedModel):
|
| 96 |
-
"""
|
| 97 |
-
GraniteVisionEmb model implementation.
|
| 98 |
-
"""
|
| 99 |
-
|
| 100 |
-
main_input_name: ClassVar[str] = "doc_input_ids" # transformers-related
|
| 101 |
-
config_class = GraniteVisionEmbConfig
|
| 102 |
-
|
| 103 |
-
def __init__(self, config: GraniteVisionEmbConfig):
|
| 104 |
-
super().__init__(config=config)
|
| 105 |
-
|
| 106 |
-
model = LlavaNextWithCustomPacking(config=config)
|
| 107 |
-
if model.language_model._tied_weights_keys is not None:
|
| 108 |
-
self._tied_weights_keys = [f"model.language_model.{k}" for k in model.language_model._tied_weights_keys]
|
| 109 |
-
self.model = model
|
| 110 |
-
|
| 111 |
-
self.dim = 128
|
| 112 |
-
self.custom_text_proj = nn.Linear(self.model.config.text_config.hidden_size, self.dim)
|
| 113 |
-
|
| 114 |
-
self.post_init()
|
| 115 |
-
|
| 116 |
-
def forward(self, *args, **kwargs) -> torch.Tensor:
|
| 117 |
-
# Delete output_hidden_states from kwargs
|
| 118 |
-
kwargs.pop("output_hidden_states", None)
|
| 119 |
-
if "pixel_values" in kwargs:
|
| 120 |
-
kwargs["pixel_values"] = kwargs["pixel_values"].to(dtype=self.dtype)
|
| 121 |
-
|
| 122 |
-
outputs = self.model(*args, output_hidden_states=True, **kwargs) # (batch_size, sequence_length, hidden_size)
|
| 123 |
-
last_hidden_states = outputs.hidden_states[-1] # (batch_size, sequence_length, hidden_size)
|
| 124 |
-
|
| 125 |
-
attention_mask = kwargs["attention_mask"]
|
| 126 |
-
if "pixel_values" in kwargs:
|
| 127 |
-
input_ids = kwargs['input_ids']
|
| 128 |
-
image_mask = (input_ids == self.config.image_token_index)
|
| 129 |
-
# inputs_embeds = last_hidden_states.masked_scatter(image_mask)
|
| 130 |
-
N, M = image_mask.shape
|
| 131 |
-
# Create an index matrix: each row is 0, 1, ..., M-1
|
| 132 |
-
idx = torch.arange(M, device=image_mask.device).expand(N, M)
|
| 133 |
-
# Replace False positions with -1 so they are ignored by topk (since all valid indices are >=0)
|
| 134 |
-
masked_idx = torch.where(image_mask, idx, torch.tensor(-1, device=image_mask.device))
|
| 135 |
-
topk_values, _ = torch.topk(masked_idx, k=729, dim=1)
|
| 136 |
-
last_k_indices, _ = torch.sort(topk_values, dim=1)
|
| 137 |
-
last_k_indices_exp = last_k_indices.unsqueeze(-1).expand(-1, -1, last_hidden_states.size(-1))
|
| 138 |
-
last_hidden_states = torch.gather(last_hidden_states, 1, last_k_indices_exp)
|
| 139 |
-
attention_mask = torch.gather(attention_mask, 1, last_k_indices)
|
| 140 |
-
|
| 141 |
-
attention_mask = attention_mask.unsqueeze(-1)
|
| 142 |
-
|
| 143 |
-
proj = self.custom_text_proj(last_hidden_states) # (batch_size, sequence_length, dim)
|
| 144 |
-
|
| 145 |
-
# L2 normalization
|
| 146 |
-
proj = proj / (proj.norm(dim=-1, keepdim=True) + 1e-8)
|
| 147 |
-
|
| 148 |
-
# proj = proj * kwargs["attention_mask"].unsqueeze(-1) # (batch_size, sequence_length, dim)
|
| 149 |
-
proj = proj * attention_mask # (batch_size, sequence_length, dim)
|
| 150 |
-
|
| 151 |
-
return proj
|
| 152 |
-
|
| 153 |
-
def get_input_embeddings(self):
|
| 154 |
-
return self.model.language_model.get_input_embeddings()
|
| 155 |
-
|
| 156 |
-
def set_input_embeddings(self, value):
|
| 157 |
-
self.model.language_model.set_input_embeddings(value)
|
| 158 |
-
|
| 159 |
-
def get_output_embeddings(self):
|
| 160 |
-
return self.model.language_model.get_output_embeddings()
|
| 161 |
-
|
| 162 |
-
def set_output_embeddings(self, new_embeddings):
|
| 163 |
-
self.model.language_model.set_output_embeddings(new_embeddings)
|
| 164 |
-
|
| 165 |
-
def set_decoder(self, decoder):
|
| 166 |
-
self.model.language_model.set_decoder(decoder)
|
| 167 |
-
|
| 168 |
-
def get_decoder(self):
|
| 169 |
-
return self.model.language_model.get_decoder()
|
| 170 |
-
|
| 171 |
-
def tie_weights(self):
|
| 172 |
-
return self.model.language_model.tie_weights()
|
| 173 |
-
|
| 174 |
-
def resize_token_embeddings(
|
| 175 |
-
self,
|
| 176 |
-
new_num_tokens: Optional[int] = None,
|
| 177 |
-
pad_to_multiple_of=None,
|
| 178 |
-
) -> nn.Embedding:
|
| 179 |
-
model_embeds = self.model.language_model.resize_token_embeddings(new_num_tokens, pad_to_multiple_of)
|
| 180 |
-
|
| 181 |
-
# Update vocab size
|
| 182 |
-
self.config.text_config.vocab_size = model_embeds.num_embeddings
|
| 183 |
-
self.config.vocab_size = model_embeds.num_embeddings
|
| 184 |
-
self.model.vocab_size = model_embeds.num_embeddings
|
| 185 |
-
|
| 186 |
-
return model_embeds
|
| 187 |
-
|
| 188 |
-
@property
|
| 189 |
-
def patch_size(self) -> int:
|
| 190 |
-
return self.model.vision_tower.config.patch_size
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
preprocessor_config.json
DELETED
|
@@ -1,137 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"crop_size": {
|
| 3 |
-
"height": 384,
|
| 4 |
-
"width": 384
|
| 5 |
-
},
|
| 6 |
-
"default_to_square": false,
|
| 7 |
-
"do_center_crop": true,
|
| 8 |
-
"do_convert_rgb": null,
|
| 9 |
-
"do_normalize": true,
|
| 10 |
-
"do_pad": true,
|
| 11 |
-
"do_rescale": true,
|
| 12 |
-
"do_resize": true,
|
| 13 |
-
"image_grid_pinpoints": [
|
| 14 |
-
[
|
| 15 |
-
384,
|
| 16 |
-
768
|
| 17 |
-
],
|
| 18 |
-
[
|
| 19 |
-
384,
|
| 20 |
-
1152
|
| 21 |
-
],
|
| 22 |
-
[
|
| 23 |
-
384,
|
| 24 |
-
1536
|
| 25 |
-
],
|
| 26 |
-
[
|
| 27 |
-
384,
|
| 28 |
-
1920
|
| 29 |
-
],
|
| 30 |
-
[
|
| 31 |
-
384,
|
| 32 |
-
2304
|
| 33 |
-
],
|
| 34 |
-
[
|
| 35 |
-
384,
|
| 36 |
-
2688
|
| 37 |
-
],
|
| 38 |
-
[
|
| 39 |
-
384,
|
| 40 |
-
3072
|
| 41 |
-
],
|
| 42 |
-
[
|
| 43 |
-
384,
|
| 44 |
-
3456
|
| 45 |
-
],
|
| 46 |
-
[
|
| 47 |
-
384,
|
| 48 |
-
3840
|
| 49 |
-
],
|
| 50 |
-
[
|
| 51 |
-
768,
|
| 52 |
-
384
|
| 53 |
-
],
|
| 54 |
-
[
|
| 55 |
-
768,
|
| 56 |
-
768
|
| 57 |
-
],
|
| 58 |
-
[
|
| 59 |
-
768,
|
| 60 |
-
1152
|
| 61 |
-
],
|
| 62 |
-
[
|
| 63 |
-
768,
|
| 64 |
-
1536
|
| 65 |
-
],
|
| 66 |
-
[
|
| 67 |
-
768,
|
| 68 |
-
1920
|
| 69 |
-
],
|
| 70 |
-
[
|
| 71 |
-
1152,
|
| 72 |
-
384
|
| 73 |
-
],
|
| 74 |
-
[
|
| 75 |
-
1152,
|
| 76 |
-
768
|
| 77 |
-
],
|
| 78 |
-
[
|
| 79 |
-
1152,
|
| 80 |
-
1152
|
| 81 |
-
],
|
| 82 |
-
[
|
| 83 |
-
1536,
|
| 84 |
-
384
|
| 85 |
-
],
|
| 86 |
-
[
|
| 87 |
-
1536,
|
| 88 |
-
768
|
| 89 |
-
],
|
| 90 |
-
[
|
| 91 |
-
1920,
|
| 92 |
-
384
|
| 93 |
-
],
|
| 94 |
-
[
|
| 95 |
-
1920,
|
| 96 |
-
768
|
| 97 |
-
],
|
| 98 |
-
[
|
| 99 |
-
2304,
|
| 100 |
-
384
|
| 101 |
-
],
|
| 102 |
-
[
|
| 103 |
-
2688,
|
| 104 |
-
384
|
| 105 |
-
],
|
| 106 |
-
[
|
| 107 |
-
3072,
|
| 108 |
-
384
|
| 109 |
-
],
|
| 110 |
-
[
|
| 111 |
-
3456,
|
| 112 |
-
384
|
| 113 |
-
],
|
| 114 |
-
[
|
| 115 |
-
3840,
|
| 116 |
-
384
|
| 117 |
-
]
|
| 118 |
-
],
|
| 119 |
-
"image_mean": [
|
| 120 |
-
0.5,
|
| 121 |
-
0.5,
|
| 122 |
-
0.5
|
| 123 |
-
],
|
| 124 |
-
"image_processor_type": "LlavaNextImageProcessor",
|
| 125 |
-
"image_std": [
|
| 126 |
-
0.5,
|
| 127 |
-
0.5,
|
| 128 |
-
0.5
|
| 129 |
-
],
|
| 130 |
-
"processor_class": "GraniteVisionEmbProcessor",
|
| 131 |
-
"resample": 3,
|
| 132 |
-
"rescale_factor": 0.00392156862745098,
|
| 133 |
-
"size": {
|
| 134 |
-
"height": 384,
|
| 135 |
-
"width": 384
|
| 136 |
-
}
|
| 137 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
processing_granite_vision_embedding.py
DELETED
|
@@ -1,439 +0,0 @@
|
|
| 1 |
-
import math
|
| 2 |
-
from typing import ClassVar, List, Optional, Tuple, Union
|
| 3 |
-
|
| 4 |
-
import torch
|
| 5 |
-
from PIL import Image, ImageOps
|
| 6 |
-
from transformers import BatchFeature, LlavaNextProcessor
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
def round_by_factor(number: float, factor: int) -> int:
|
| 10 |
-
"""Returns the closest integer to 'number' that is divisible by 'factor'."""
|
| 11 |
-
return round(number / factor) * factor
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
def ceil_by_factor(number: float, factor: int) -> int:
|
| 15 |
-
"""Returns the smallest integer greater than or equal to 'number' that is divisible by 'factor'."""
|
| 16 |
-
return math.ceil(number / factor) * factor
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
def floor_by_factor(number: float, factor: int) -> int:
|
| 20 |
-
"""Returns the largest integer less than or equal to 'number' that is divisible by 'factor'."""
|
| 21 |
-
return math.floor(number / factor) * factor
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
class GraniteVisionEmbProcessor(LlavaNextProcessor):
|
| 25 |
-
"""
|
| 26 |
-
Processor for GraniteVisionEmb.
|
| 27 |
-
"""
|
| 28 |
-
|
| 29 |
-
visual_prompt_prefix: ClassVar[str] = "<|user|>\n<image>\nDescribe the image.\n"
|
| 30 |
-
system_message: ClassVar[
|
| 31 |
-
str] = "A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions."
|
| 32 |
-
query_prefix: ClassVar[str] = "Query: "
|
| 33 |
-
query_start: ClassVar[str] = "<|user|>\n"
|
| 34 |
-
|
| 35 |
-
def __init__(self, *args, **kwargs):
|
| 36 |
-
super().__init__(*args, **kwargs)
|
| 37 |
-
self.factor = 14
|
| 38 |
-
self.min_size = 384
|
| 39 |
-
self.max_size = 384 * 2
|
| 40 |
-
self.suffix_len = 10
|
| 41 |
-
self.patch_size = 14
|
| 42 |
-
|
| 43 |
-
@property
|
| 44 |
-
def query_augmentation_token(self) -> str:
|
| 45 |
-
"""
|
| 46 |
-
Return the query augmentation token.
|
| 47 |
-
Query augmentation buffers are used as reasoning buffers during inference.
|
| 48 |
-
"""
|
| 49 |
-
return self.tokenizer.pad_token
|
| 50 |
-
|
| 51 |
-
@staticmethod
|
| 52 |
-
def smart_resize_helper(
|
| 53 |
-
width: int,
|
| 54 |
-
height: int,
|
| 55 |
-
factor: int,
|
| 56 |
-
min_size: int,
|
| 57 |
-
max_size: int
|
| 58 |
-
) -> Tuple[int, int]:
|
| 59 |
-
"""
|
| 60 |
-
Returns the resized image dimensions such that:
|
| 61 |
-
1. The smaller dimension is set to 'min_size'.
|
| 62 |
-
2. The larger dimension is scaled proportionally to maintain aspect ratio.
|
| 63 |
-
3. If the larger dimension exceeds 'max_size', it is clipped to 'max_size',
|
| 64 |
-
and the smaller dimension is adjusted accordingly to maintain aspect ratio.
|
| 65 |
-
4. Both dimensions are divisible by 'factor'.
|
| 66 |
-
"""
|
| 67 |
-
|
| 68 |
-
# Determine scale factor based on min_size
|
| 69 |
-
if height < width:
|
| 70 |
-
scale_factor = min_size / height
|
| 71 |
-
else:
|
| 72 |
-
scale_factor = min_size / width
|
| 73 |
-
|
| 74 |
-
new_width = round(width * scale_factor)
|
| 75 |
-
new_height = round(height * scale_factor)
|
| 76 |
-
|
| 77 |
-
# If the longer dimension exceeds max_size, adjust accordingly
|
| 78 |
-
if max(new_width, new_height) > max_size:
|
| 79 |
-
clip_factor = max_size / max(new_width, new_height)
|
| 80 |
-
new_width = round(new_width * clip_factor)
|
| 81 |
-
new_height = round(new_height * clip_factor)
|
| 82 |
-
|
| 83 |
-
# Ensure dimensions are divisible by factor
|
| 84 |
-
# new_width = round_by_factor(new_width, factor)
|
| 85 |
-
# new_height = round_by_factor(new_height, factor)
|
| 86 |
-
|
| 87 |
-
return new_width, new_height
|
| 88 |
-
|
| 89 |
-
@staticmethod
|
| 90 |
-
def pad_image_center(image: Image.Image,
|
| 91 |
-
target_width: int,
|
| 92 |
-
target_height: int,
|
| 93 |
-
fill_color=(0, 0, 0)) -> Image.Image:
|
| 94 |
-
"""
|
| 95 |
-
Pads the given image to be centered within the target dimensions.
|
| 96 |
-
|
| 97 |
-
:param image: PIL Image to be padded.
|
| 98 |
-
:param target_width: The desired width after padding.
|
| 99 |
-
:param target_height: The desired height after padding.
|
| 100 |
-
:param fill_color: Background color (default is black).
|
| 101 |
-
:return: Padded image with centered content.
|
| 102 |
-
"""
|
| 103 |
-
|
| 104 |
-
# Get original image size
|
| 105 |
-
img_width, img_height = image.size
|
| 106 |
-
|
| 107 |
-
# Compute padding values
|
| 108 |
-
pad_left = (target_width - img_width) // 2
|
| 109 |
-
pad_top = (target_height - img_height) // 2
|
| 110 |
-
pad_right = target_width - img_width - pad_left
|
| 111 |
-
pad_bottom = target_height - img_height - pad_top
|
| 112 |
-
|
| 113 |
-
# Apply padding
|
| 114 |
-
padded_image = ImageOps.expand(image, (pad_left, pad_top, pad_right, pad_bottom), fill_color).convert("RGB")
|
| 115 |
-
|
| 116 |
-
return padded_image
|
| 117 |
-
|
| 118 |
-
def smart_resize(self, image: Image.Image) -> Image.Image:
|
| 119 |
-
"""
|
| 120 |
-
Resize and convert the image to the required format.
|
| 121 |
-
"""
|
| 122 |
-
image_size = image.size
|
| 123 |
-
resized_height, resized_width = self.smart_resize_helper(
|
| 124 |
-
width=image_size[0],
|
| 125 |
-
height=image_size[1],
|
| 126 |
-
factor=self.factor,
|
| 127 |
-
min_size=self.min_size,
|
| 128 |
-
max_size=self.max_size
|
| 129 |
-
)
|
| 130 |
-
return image.convert("RGB").resize((resized_width, resized_height))
|
| 131 |
-
|
| 132 |
-
def smart_resize_and_pad(self, image: Image.Image) -> Image.Image:
|
| 133 |
-
"""
|
| 134 |
-
Resize and pad the image to the required format.
|
| 135 |
-
"""
|
| 136 |
-
return self.resize_and_pad_centered_to_long_side(
|
| 137 |
-
image=image,
|
| 138 |
-
factor=self.factor,
|
| 139 |
-
min_size=self.min_size,
|
| 140 |
-
max_size=self.max_size,
|
| 141 |
-
fill_color=0
|
| 142 |
-
)
|
| 143 |
-
|
| 144 |
-
def resize_and_pad_centered_to_long_side(
|
| 145 |
-
self,
|
| 146 |
-
image: Image.Image,
|
| 147 |
-
factor: int,
|
| 148 |
-
min_size: int,
|
| 149 |
-
max_size: int,
|
| 150 |
-
fill_color=0
|
| 151 |
-
) -> Image.Image:
|
| 152 |
-
"""
|
| 153 |
-
Resizes and pads an image such that:
|
| 154 |
-
- The long side is set to `max_size`.
|
| 155 |
-
- The short side is scaled proportionally but not below `min_size`.
|
| 156 |
-
- The image is centered within the final padded area.
|
| 157 |
-
|
| 158 |
-
:param image: PIL Image
|
| 159 |
-
:param factor: Factor to make dimensions divisible by
|
| 160 |
-
:param min_size: Minimum allowed size for the short side
|
| 161 |
-
:param max_size: Target size for the long side
|
| 162 |
-
:param fill_color: Background padding color (default black)
|
| 163 |
-
:return: Resized and padded image
|
| 164 |
-
"""
|
| 165 |
-
|
| 166 |
-
# Get original size
|
| 167 |
-
width, height = image.size
|
| 168 |
-
|
| 169 |
-
if min_size == -1 or max_size == -1:
|
| 170 |
-
return image.convert("RGB")
|
| 171 |
-
|
| 172 |
-
# Step 1: scale long side to max_size, keep aspect ratio
|
| 173 |
-
if width > height:
|
| 174 |
-
scale_factor = max_size / width
|
| 175 |
-
target_width = max_size
|
| 176 |
-
max_scale_factor = max(min_size / height, scale_factor)
|
| 177 |
-
target_height = round(height * max_scale_factor)
|
| 178 |
-
else:
|
| 179 |
-
scale_factor = max_size / height
|
| 180 |
-
target_height = max_size
|
| 181 |
-
max_scale_factor = max(min_size / width, scale_factor)
|
| 182 |
-
target_width = round(width * max_scale_factor)
|
| 183 |
-
|
| 184 |
-
# Resize the image
|
| 185 |
-
resized_image = image.resize((target_width, target_height), Image.LANCZOS)
|
| 186 |
-
final_image = resized_image.convert("RGB")
|
| 187 |
-
|
| 188 |
-
return final_image
|
| 189 |
-
|
| 190 |
-
def resize_and_pad_centered(self,
|
| 191 |
-
image: Image.Image,
|
| 192 |
-
factor: int,
|
| 193 |
-
min_size: int,
|
| 194 |
-
max_size: int,
|
| 195 |
-
fill_color=0
|
| 196 |
-
) -> Image.Image:
|
| 197 |
-
"""
|
| 198 |
-
Resizes and pads an image such that:
|
| 199 |
-
- The short side is set to `min_size`.
|
| 200 |
-
- The long side is scaled proportionally but clipped to `max_size`.
|
| 201 |
-
- The image is centered within the final padded area.
|
| 202 |
-
|
| 203 |
-
:param image: PIL Image
|
| 204 |
-
:param factor: Factor to make dimensions divisible by
|
| 205 |
-
:param min_size: Minimum size for the short side
|
| 206 |
-
:param max_size: Maximum allowed size for the long side
|
| 207 |
-
:param fill_color: Background padding color (default black)
|
| 208 |
-
:return: Resized and padded image
|
| 209 |
-
"""
|
| 210 |
-
|
| 211 |
-
# Get original size
|
| 212 |
-
width, height = image.size
|
| 213 |
-
|
| 214 |
-
if min_size == -1 or max_size == -1:
|
| 215 |
-
return image.convert("RGB")
|
| 216 |
-
|
| 217 |
-
# Determine scale factor based on the short side (min_size)
|
| 218 |
-
if width < height:
|
| 219 |
-
scale_factor = min_size / width
|
| 220 |
-
target_width = min_size
|
| 221 |
-
max_scale_factor = min(max_size / height, scale_factor)
|
| 222 |
-
target_height = round(height * max_scale_factor)
|
| 223 |
-
else:
|
| 224 |
-
scale_factor = min_size / height
|
| 225 |
-
target_height = min_size
|
| 226 |
-
max_scale_factor = min(max_size / width, scale_factor)
|
| 227 |
-
target_width = round(width * max_scale_factor)
|
| 228 |
-
|
| 229 |
-
# Ensure the longer side does not exceed max_size
|
| 230 |
-
# if max(target_width, target_height) > max_size:
|
| 231 |
-
# clip_factor = max_size / max(target_width, target_height)
|
| 232 |
-
# target_width = round(target_width * clip_factor)
|
| 233 |
-
# target_height = round(target_height * clip_factor)
|
| 234 |
-
|
| 235 |
-
# Ensure dimensions are divisible by factor
|
| 236 |
-
# target_width = round_by_factor(target_width, factor)
|
| 237 |
-
# target_height = round_by_factor(target_height, factor)
|
| 238 |
-
|
| 239 |
-
# Resize the image
|
| 240 |
-
resized_image = image.resize((target_width, target_height), Image.LANCZOS)
|
| 241 |
-
|
| 242 |
-
# Determine final padded dimensions (aligned to short side)
|
| 243 |
-
if width < height:
|
| 244 |
-
final_width, final_height = min_size, max_size
|
| 245 |
-
else:
|
| 246 |
-
final_width, final_height = max_size, min_size
|
| 247 |
-
|
| 248 |
-
# Compute padding to center the image
|
| 249 |
-
pad_left = (final_width - target_width) // 2
|
| 250 |
-
pad_top = (final_height - target_height) // 2
|
| 251 |
-
pad_right = final_width - target_width - pad_left
|
| 252 |
-
pad_bottom = final_height - target_height - pad_top
|
| 253 |
-
|
| 254 |
-
# Apply centered padding
|
| 255 |
-
# final_image = ImageOps.expand(resized_image, (pad_left, pad_top, pad_right, pad_bottom), fill_color).convert("RGB")
|
| 256 |
-
final_image = resized_image.convert("RGB")
|
| 257 |
-
|
| 258 |
-
return final_image
|
| 259 |
-
|
| 260 |
-
def format_data(self, question, image):
|
| 261 |
-
return [
|
| 262 |
-
{
|
| 263 |
-
"role": "system",
|
| 264 |
-
"content": [{"type": "text", "text": self.system_message}],
|
| 265 |
-
},
|
| 266 |
-
{
|
| 267 |
-
"role": "user",
|
| 268 |
-
"content": [
|
| 269 |
-
{
|
| 270 |
-
"type": "image",
|
| 271 |
-
"image": image,
|
| 272 |
-
},
|
| 273 |
-
{
|
| 274 |
-
"type": "text",
|
| 275 |
-
"text": question,
|
| 276 |
-
},
|
| 277 |
-
],
|
| 278 |
-
}
|
| 279 |
-
]
|
| 280 |
-
|
| 281 |
-
def format_data_wo_role(self, question, image=None):
|
| 282 |
-
return [
|
| 283 |
-
{
|
| 284 |
-
"role": "user",
|
| 285 |
-
"content": [
|
| 286 |
-
{
|
| 287 |
-
"type": "image",
|
| 288 |
-
"image": image,
|
| 289 |
-
},
|
| 290 |
-
{
|
| 291 |
-
"type": "text",
|
| 292 |
-
"text": question,
|
| 293 |
-
},
|
| 294 |
-
],
|
| 295 |
-
}
|
| 296 |
-
]
|
| 297 |
-
|
| 298 |
-
def process_images(
|
| 299 |
-
self,
|
| 300 |
-
images: List[Image.Image],
|
| 301 |
-
) -> BatchFeature:
|
| 302 |
-
"""
|
| 303 |
-
Process images.
|
| 304 |
-
"""
|
| 305 |
-
# texts_doc = [self.apply_chat_template(self.format_data_wo_role(self.visual_prompt_prefix, img),tokenize=False ) for img in images]
|
| 306 |
-
texts_doc = [self.visual_prompt_prefix for _ in images]
|
| 307 |
-
images = [self.smart_resize_and_pad(image) for image in images]
|
| 308 |
-
|
| 309 |
-
batch_doc = self(
|
| 310 |
-
text=texts_doc,
|
| 311 |
-
images=images,
|
| 312 |
-
return_tensors="pt",
|
| 313 |
-
padding="longest",
|
| 314 |
-
)
|
| 315 |
-
return batch_doc
|
| 316 |
-
|
| 317 |
-
def process_queries(self, queries, max_length=2048, suffix=None):
|
| 318 |
-
if suffix is None:
|
| 319 |
-
suffix = self.query_augmentation_token * self.suffix_len
|
| 320 |
-
|
| 321 |
-
processed = []
|
| 322 |
-
for q in queries:
|
| 323 |
-
q = self.query_start + self.query_prefix + q + ' ' + q
|
| 324 |
-
q += suffix + "\n"
|
| 325 |
-
processed.append(q)
|
| 326 |
-
|
| 327 |
-
return self(
|
| 328 |
-
text=processed,
|
| 329 |
-
images=None,
|
| 330 |
-
return_tensors="pt",
|
| 331 |
-
padding="longest",
|
| 332 |
-
truncation=True,
|
| 333 |
-
max_length=max_length,
|
| 334 |
-
)
|
| 335 |
-
|
| 336 |
-
def score(
|
| 337 |
-
self,
|
| 338 |
-
qs: List[torch.Tensor],
|
| 339 |
-
ps: List[torch.Tensor],
|
| 340 |
-
device: Optional[Union[str, torch.device]] = None,
|
| 341 |
-
**kwargs,
|
| 342 |
-
) -> torch.Tensor:
|
| 343 |
-
"""
|
| 344 |
-
Compute the MaxSim score (ColBERT-like) for the given multi-vector query and passage embeddings.
|
| 345 |
-
"""
|
| 346 |
-
return self.score_multi_vector(qs, ps, device=device, **kwargs)
|
| 347 |
-
|
| 348 |
-
def get_n_patches(
|
| 349 |
-
self,
|
| 350 |
-
image_size: Tuple[int, int],
|
| 351 |
-
patch_size: int,
|
| 352 |
-
) -> Tuple[int, int]:
|
| 353 |
-
n_patches_x = self.image_processor.size["width"] // patch_size
|
| 354 |
-
n_patches_y = self.image_processor.size["height"] // patch_size
|
| 355 |
-
|
| 356 |
-
return n_patches_x, n_patches_y
|
| 357 |
-
|
| 358 |
-
def get_image_mask(self, batch_images: BatchFeature) -> torch.Tensor:
|
| 359 |
-
return batch_images.input_ids == self.image_token_id
|
| 360 |
-
|
| 361 |
-
@staticmethod
|
| 362 |
-
def score_single_vector(
|
| 363 |
-
qs: List[torch.Tensor],
|
| 364 |
-
ps: List[torch.Tensor],
|
| 365 |
-
device: Optional[Union[str, torch.device]] = None,
|
| 366 |
-
) -> torch.Tensor:
|
| 367 |
-
"""
|
| 368 |
-
Compute the dot product score for the given single-vector query and passage embeddings.
|
| 369 |
-
"""
|
| 370 |
-
|
| 371 |
-
if len(qs) == 0:
|
| 372 |
-
raise ValueError("No queries provided")
|
| 373 |
-
if len(ps) == 0:
|
| 374 |
-
raise ValueError("No passages provided")
|
| 375 |
-
|
| 376 |
-
qs_stacked = torch.stack(qs).to(device)
|
| 377 |
-
ps_stacked = torch.stack(ps).to(device)
|
| 378 |
-
|
| 379 |
-
scores = torch.einsum("bd,cd->bc", qs_stacked, ps_stacked)
|
| 380 |
-
assert scores.shape[0] == len(qs), f"Expected {len(qs)} scores, got {scores.shape[0]}"
|
| 381 |
-
|
| 382 |
-
scores = scores.to(torch.float32)
|
| 383 |
-
return scores
|
| 384 |
-
|
| 385 |
-
@staticmethod
|
| 386 |
-
def score_multi_vector(
|
| 387 |
-
qs: Union[torch.Tensor, List[torch.Tensor]],
|
| 388 |
-
ps: Union[torch.Tensor, List[torch.Tensor]],
|
| 389 |
-
batch_size: int = 128,
|
| 390 |
-
device: Optional[Union[str, torch.device]] = None,
|
| 391 |
-
) -> torch.Tensor:
|
| 392 |
-
"""
|
| 393 |
-
Compute the late-interaction/MaxSim score (ColBERT-like) for the given multi-vector
|
| 394 |
-
query embeddings (`qs`) and passage embeddings (`ps`). For us, a passage is the
|
| 395 |
-
image of a document page.
|
| 396 |
-
|
| 397 |
-
Because the embedding tensors are multi-vector and can thus have different shapes, they
|
| 398 |
-
should be fed as:
|
| 399 |
-
(1) a list of tensors, where the i-th tensor is of shape (sequence_length_i, embedding_dim)
|
| 400 |
-
(2) a single tensor of shape (n_passages, max_sequence_length, embedding_dim) -> usually
|
| 401 |
-
obtained by padding the list of tensors.
|
| 402 |
-
|
| 403 |
-
Args:
|
| 404 |
-
qs (`Union[torch.Tensor, List[torch.Tensor]`): Query embeddings.
|
| 405 |
-
ps (`Union[torch.Tensor, List[torch.Tensor]`): Passage embeddings.
|
| 406 |
-
batch_size (`int`, *optional*, defaults to 128): Batch size for computing scores.
|
| 407 |
-
device (`Union[str, torch.device]`, *optional*): Device to use for computation. If not
|
| 408 |
-
provided, uses `get_torch_device("auto")`.
|
| 409 |
-
|
| 410 |
-
Returns:
|
| 411 |
-
`torch.Tensor`: A tensor of shape `(n_queries, n_passages)` containing the scores. The score
|
| 412 |
-
tensor is saved on the "cpu" device.
|
| 413 |
-
"""
|
| 414 |
-
|
| 415 |
-
if len(qs) == 0:
|
| 416 |
-
raise ValueError("No queries provided")
|
| 417 |
-
if len(ps) == 0:
|
| 418 |
-
raise ValueError("No passages provided")
|
| 419 |
-
|
| 420 |
-
scores_list: List[torch.Tensor] = []
|
| 421 |
-
|
| 422 |
-
for i in range(0, len(qs), batch_size):
|
| 423 |
-
scores_batch = []
|
| 424 |
-
qs_batch = torch.nn.utils.rnn.pad_sequence(qs[i: i + batch_size], batch_first=True, padding_value=0).to(
|
| 425 |
-
device
|
| 426 |
-
)
|
| 427 |
-
for j in range(0, len(ps), batch_size):
|
| 428 |
-
ps_batch = torch.nn.utils.rnn.pad_sequence(
|
| 429 |
-
ps[j: j + batch_size], batch_first=True, padding_value=0
|
| 430 |
-
).to(device)
|
| 431 |
-
scores_batch.append(torch.einsum("bnd,csd->bcns", qs_batch, ps_batch).max(dim=3)[0].sum(dim=2))
|
| 432 |
-
scores_batch = torch.cat(scores_batch, dim=1).cpu()
|
| 433 |
-
scores_list.append(scores_batch)
|
| 434 |
-
|
| 435 |
-
scores = torch.cat(scores_list, dim=0)
|
| 436 |
-
assert scores.shape[0] == len(qs), f"Expected {len(qs)} scores, got {scores.shape[0]}"
|
| 437 |
-
|
| 438 |
-
scores = scores.to(torch.float32)
|
| 439 |
-
return scores
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
processor_config.json
DELETED
|
@@ -1,6 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"processor_class": "GraniteVisionEmbProcessor",
|
| 3 |
-
"auto_map": {
|
| 4 |
-
"AutoProcessor": "processing_granite_vision_embedding.GraniteVisionEmbProcessor"
|
| 5 |
-
}
|
| 6 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
special_tokens_map.json
DELETED
|
@@ -1,35 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"additional_special_tokens": [
|
| 3 |
-
"<|start_of_role|>",
|
| 4 |
-
"<|end_of_role|>",
|
| 5 |
-
"<|tool_call|>"
|
| 6 |
-
],
|
| 7 |
-
"bos_token": {
|
| 8 |
-
"content": "<|end_of_text|>",
|
| 9 |
-
"lstrip": false,
|
| 10 |
-
"normalized": false,
|
| 11 |
-
"rstrip": false,
|
| 12 |
-
"single_word": false
|
| 13 |
-
},
|
| 14 |
-
"eos_token": {
|
| 15 |
-
"content": "<|end_of_text|>",
|
| 16 |
-
"lstrip": false,
|
| 17 |
-
"normalized": false,
|
| 18 |
-
"rstrip": false,
|
| 19 |
-
"single_word": false
|
| 20 |
-
},
|
| 21 |
-
"pad_token": {
|
| 22 |
-
"content": "<|end_of_text|>",
|
| 23 |
-
"lstrip": false,
|
| 24 |
-
"normalized": false,
|
| 25 |
-
"rstrip": false,
|
| 26 |
-
"single_word": false
|
| 27 |
-
},
|
| 28 |
-
"unk_token": {
|
| 29 |
-
"content": "<|end_of_text|>",
|
| 30 |
-
"lstrip": false,
|
| 31 |
-
"normalized": false,
|
| 32 |
-
"rstrip": false,
|
| 33 |
-
"single_word": false
|
| 34 |
-
}
|
| 35 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
tokenizer.json
DELETED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
DELETED
|
@@ -1,208 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"add_bos_token": false,
|
| 3 |
-
"add_prefix_space": false,
|
| 4 |
-
"added_tokens_decoder": {
|
| 5 |
-
"0": {
|
| 6 |
-
"content": "<|end_of_text|>",
|
| 7 |
-
"lstrip": false,
|
| 8 |
-
"normalized": false,
|
| 9 |
-
"rstrip": false,
|
| 10 |
-
"single_word": false,
|
| 11 |
-
"special": true
|
| 12 |
-
},
|
| 13 |
-
"1": {
|
| 14 |
-
"content": "<fim_prefix>",
|
| 15 |
-
"lstrip": false,
|
| 16 |
-
"normalized": false,
|
| 17 |
-
"rstrip": false,
|
| 18 |
-
"single_word": false,
|
| 19 |
-
"special": true
|
| 20 |
-
},
|
| 21 |
-
"2": {
|
| 22 |
-
"content": "<fim_middle>",
|
| 23 |
-
"lstrip": false,
|
| 24 |
-
"normalized": false,
|
| 25 |
-
"rstrip": false,
|
| 26 |
-
"single_word": false,
|
| 27 |
-
"special": true
|
| 28 |
-
},
|
| 29 |
-
"3": {
|
| 30 |
-
"content": "<fim_suffix>",
|
| 31 |
-
"lstrip": false,
|
| 32 |
-
"normalized": false,
|
| 33 |
-
"rstrip": false,
|
| 34 |
-
"single_word": false,
|
| 35 |
-
"special": true
|
| 36 |
-
},
|
| 37 |
-
"4": {
|
| 38 |
-
"content": "<fim_pad>",
|
| 39 |
-
"lstrip": false,
|
| 40 |
-
"normalized": false,
|
| 41 |
-
"rstrip": false,
|
| 42 |
-
"single_word": false,
|
| 43 |
-
"special": true
|
| 44 |
-
},
|
| 45 |
-
"5": {
|
| 46 |
-
"content": "<filename>",
|
| 47 |
-
"lstrip": false,
|
| 48 |
-
"normalized": false,
|
| 49 |
-
"rstrip": false,
|
| 50 |
-
"single_word": false,
|
| 51 |
-
"special": true
|
| 52 |
-
},
|
| 53 |
-
"6": {
|
| 54 |
-
"content": "<gh_stars>",
|
| 55 |
-
"lstrip": false,
|
| 56 |
-
"normalized": false,
|
| 57 |
-
"rstrip": false,
|
| 58 |
-
"single_word": false,
|
| 59 |
-
"special": true
|
| 60 |
-
},
|
| 61 |
-
"7": {
|
| 62 |
-
"content": "<issue_start>",
|
| 63 |
-
"lstrip": false,
|
| 64 |
-
"normalized": false,
|
| 65 |
-
"rstrip": false,
|
| 66 |
-
"single_word": false,
|
| 67 |
-
"special": true
|
| 68 |
-
},
|
| 69 |
-
"8": {
|
| 70 |
-
"content": "<issue_comment>",
|
| 71 |
-
"lstrip": false,
|
| 72 |
-
"normalized": false,
|
| 73 |
-
"rstrip": false,
|
| 74 |
-
"single_word": false,
|
| 75 |
-
"special": true
|
| 76 |
-
},
|
| 77 |
-
"9": {
|
| 78 |
-
"content": "<issue_closed>",
|
| 79 |
-
"lstrip": false,
|
| 80 |
-
"normalized": false,
|
| 81 |
-
"rstrip": false,
|
| 82 |
-
"single_word": false,
|
| 83 |
-
"special": true
|
| 84 |
-
},
|
| 85 |
-
"10": {
|
| 86 |
-
"content": "<jupyter_start>",
|
| 87 |
-
"lstrip": false,
|
| 88 |
-
"normalized": false,
|
| 89 |
-
"rstrip": false,
|
| 90 |
-
"single_word": false,
|
| 91 |
-
"special": true
|
| 92 |
-
},
|
| 93 |
-
"11": {
|
| 94 |
-
"content": "<jupyter_text>",
|
| 95 |
-
"lstrip": false,
|
| 96 |
-
"normalized": false,
|
| 97 |
-
"rstrip": false,
|
| 98 |
-
"single_word": false,
|
| 99 |
-
"special": true
|
| 100 |
-
},
|
| 101 |
-
"12": {
|
| 102 |
-
"content": "<jupyter_code>",
|
| 103 |
-
"lstrip": false,
|
| 104 |
-
"normalized": false,
|
| 105 |
-
"rstrip": false,
|
| 106 |
-
"single_word": false,
|
| 107 |
-
"special": true
|
| 108 |
-
},
|
| 109 |
-
"13": {
|
| 110 |
-
"content": "<jupyter_output>",
|
| 111 |
-
"lstrip": false,
|
| 112 |
-
"normalized": false,
|
| 113 |
-
"rstrip": false,
|
| 114 |
-
"single_word": false,
|
| 115 |
-
"special": true
|
| 116 |
-
},
|
| 117 |
-
"14": {
|
| 118 |
-
"content": "<empty_output>",
|
| 119 |
-
"lstrip": false,
|
| 120 |
-
"normalized": false,
|
| 121 |
-
"rstrip": false,
|
| 122 |
-
"single_word": false,
|
| 123 |
-
"special": true
|
| 124 |
-
},
|
| 125 |
-
"15": {
|
| 126 |
-
"content": "<commit_before>",
|
| 127 |
-
"lstrip": false,
|
| 128 |
-
"normalized": false,
|
| 129 |
-
"rstrip": false,
|
| 130 |
-
"single_word": false,
|
| 131 |
-
"special": true
|
| 132 |
-
},
|
| 133 |
-
"16": {
|
| 134 |
-
"content": "<commit_msg>",
|
| 135 |
-
"lstrip": false,
|
| 136 |
-
"normalized": false,
|
| 137 |
-
"rstrip": false,
|
| 138 |
-
"single_word": false,
|
| 139 |
-
"special": true
|
| 140 |
-
},
|
| 141 |
-
"17": {
|
| 142 |
-
"content": "<commit_after>",
|
| 143 |
-
"lstrip": false,
|
| 144 |
-
"normalized": false,
|
| 145 |
-
"rstrip": false,
|
| 146 |
-
"single_word": false,
|
| 147 |
-
"special": true
|
| 148 |
-
},
|
| 149 |
-
"18": {
|
| 150 |
-
"content": "<reponame>",
|
| 151 |
-
"lstrip": false,
|
| 152 |
-
"normalized": false,
|
| 153 |
-
"rstrip": false,
|
| 154 |
-
"single_word": false,
|
| 155 |
-
"special": true
|
| 156 |
-
},
|
| 157 |
-
"49152": {
|
| 158 |
-
"content": "<|start_of_role|>",
|
| 159 |
-
"lstrip": false,
|
| 160 |
-
"normalized": false,
|
| 161 |
-
"rstrip": false,
|
| 162 |
-
"single_word": false,
|
| 163 |
-
"special": true
|
| 164 |
-
},
|
| 165 |
-
"49153": {
|
| 166 |
-
"content": "<|end_of_role|>",
|
| 167 |
-
"lstrip": false,
|
| 168 |
-
"normalized": false,
|
| 169 |
-
"rstrip": false,
|
| 170 |
-
"single_word": false,
|
| 171 |
-
"special": true
|
| 172 |
-
},
|
| 173 |
-
"49154": {
|
| 174 |
-
"content": "<|tool_call|>",
|
| 175 |
-
"lstrip": false,
|
| 176 |
-
"normalized": false,
|
| 177 |
-
"rstrip": false,
|
| 178 |
-
"single_word": false,
|
| 179 |
-
"special": true
|
| 180 |
-
},
|
| 181 |
-
"49155": {
|
| 182 |
-
"content": "<image>",
|
| 183 |
-
"lstrip": false,
|
| 184 |
-
"normalized": false,
|
| 185 |
-
"rstrip": false,
|
| 186 |
-
"single_word": false,
|
| 187 |
-
"special": true
|
| 188 |
-
}
|
| 189 |
-
},
|
| 190 |
-
"additional_special_tokens": [
|
| 191 |
-
"<|start_of_role|>",
|
| 192 |
-
"<|end_of_role|>",
|
| 193 |
-
"<|tool_call|>"
|
| 194 |
-
],
|
| 195 |
-
"bos_token": "<|end_of_text|>",
|
| 196 |
-
"chat_template": "{%- if tools %}\n {{- '<|start_of_role|>available_tools<|end_of_role|>\n' }}\n {%- for tool in tools %}\n {{- tool | tojson(indent=4) }}\n {%- if not loop.last %}\n {{- '\n\n' }}\n {%- endif %}\n {%- endfor %}\n {{- '<|end_of_text|>\n' }}\n{%- endif %}\n{%- for message in messages if message['role'] == 'system'%}{% else %}<|system|>\nA chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.\n{% endfor %}{%- for message in messages %}\n {%- if message['role'] == 'system' %}\n {{- '<|system|>\n' + message['content'] + '\n' }}\n {%- elif message['role'] == 'user' %}\n {{- '<|user|>\n' + message['content'] + '\n' }}\n {%- elif message['role'] == 'assistant' %}\n {{- '<|assistant|>\n' + message['content'] + '<|end_of_text|>' }}\n {%- elif message['role'] == 'assistant_tool_call' %}\n {{- '<|start_of_role|>assistant<|end_of_role|><|tool_call|>' + message['content'] + '<|end_of_text|>\n' }}\n {%- elif message['role'] == 'tool_response' %}\n {{- '<|start_of_role|>tool_response<|end_of_role|>' + message['content'] + '<|end_of_text|>\n' }}\n {%- endif %}\n {%- if loop.last and add_generation_prompt %}\n {{- '<|assistant|>\n' }}\n {%- endif %}\n{%- endfor %}",
|
| 197 |
-
"clean_up_tokenization_spaces": true,
|
| 198 |
-
"do_image_splitting": false,
|
| 199 |
-
"eos_token": "<|end_of_text|>",
|
| 200 |
-
"errors": "replace",
|
| 201 |
-
"extra_special_tokens": {},
|
| 202 |
-
"model_max_length": 131072,
|
| 203 |
-
"pad_token": "<|end_of_text|>",
|
| 204 |
-
"padding_side": "right",
|
| 205 |
-
"tokenizer_class": "GPT2Tokenizer",
|
| 206 |
-
"unk_token": "<|end_of_text|>",
|
| 207 |
-
"vocab_size": 49152
|
| 208 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
vocab.json
DELETED
|
The diff for this file is too large to render.
See raw diff
|
|
|