first upload
Browse files- .gitattributes +3 -0
- README.md +234 -0
- config.json +104 -0
- generation_config.json +13 -0
- images/MMEB-results-with-SCaR.png +3 -0
- images/MMEB-results.png +3 -0
- images/SCaR-results.png +3 -0
- images/VIRTUE-framework.jpg +3 -0
- images/example.jpg +3 -0
- preprocessor_config.json +29 -0
- pytorch_model-00001-of-00004.bin +3 -0
- pytorch_model-00002-of-00004.bin +3 -0
- pytorch_model-00003-of-00004.bin +3 -0
- pytorch_model-00004-of-00004.bin +3 -0
- pytorch_model.bin.index.json +737 -0
- sam2_reducer_and_vlm_layer.pth +3 -0
- special_tokens_map.json +31 -0
- tokenizer.json +3 -0
- tokenizer_config.json +144 -0
- vocab.json +0 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,6 @@ 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
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
*.png filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
*.jpg filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
|
@@ -1,3 +1,237 @@
|
|
| 1 |
---
|
| 2 |
license: cc-by-nc-4.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: cc-by-nc-4.0
|
| 3 |
+
datasets:
|
| 4 |
+
- TIGER-Lab/MMEB-train
|
| 5 |
+
- Sony/SCaR-Train
|
| 6 |
+
language:
|
| 7 |
+
- en
|
| 8 |
+
metrics:
|
| 9 |
+
- accuracy
|
| 10 |
+
base_model:
|
| 11 |
+
- Qwen/Qwen2-VL-7B-Instruct
|
| 12 |
+
library_name: transformers
|
| 13 |
+
tags:
|
| 14 |
+
- Embedding
|
| 15 |
---
|
| 16 |
+
|
| 17 |
+
# VIRTUE Model Card
|
| 18 |
+
|
| 19 |
+
[VIRTUE-2B-SCaR](https://huggingface.co/Sony/VIRTUE-2B-SCaR) and [VIRTUE-7B-SCaR](https://huggingface.co/Sony/VIRTUE-7B-SCaR) are the official checkpoints for the paper "[VIRTUE: Visual-Interactive Text-Image Universal Embedder](https://arxiv.org/abs/2510.00523)" that are trained with MMEB-Train and SCaR-Train.
|
| 20 |
+
VIRTUE is a visual-interactive text-image universal embedder consisting of a VLM as well as a segmentation model to enable the visual interaction modality for human interactions.
|
| 21 |
+
In addition, we introduce the SCaR benchmark ([train](https://huggingface.co/datasets/Sony/SCaR-Train), [eval](https://huggingface.co/datasets/Sony/SCaR-Eval)), composed of 1M samples for visual-interactive image-to-text retrieval, to evaluate visual-interactive embedding capabilities.
|
| 22 |
+
SCaR enables evaluation of advanced reasoning and compositional tasks in multimodal, visual-interaction-aware embedding scenarios that remain unexplored.
|
| 23 |
+
|
| 24 |
+
## Model Checkpoints
|
| 25 |
+
|
| 26 |
+
- [VIRTUE-2B-SCaR](https://huggingface.co/Sony/VIRTUE-2B-SCaR)
|
| 27 |
+
- [VIRTUE-7B-SCaR](https://huggingface.co/Sony/VIRTUE-7B-SCaR)
|
| 28 |
+
|
| 29 |
+
## Experimental Results
|
| 30 |
+
|
| 31 |
+
### MMEB
|
| 32 |
+
- Without SCaR-Train:
|
| 33 |
+
|
| 34 |
+
|
| 35 |
+
- With SCaR-Train
|
| 36 |
+
|
| 37 |
+
|
| 38 |
+
|
| 39 |
+
|
| 40 |
+
### SCaR
|
| 41 |
+
|
| 42 |
+
|
| 43 |
+
## Resources
|
| 44 |
+
- [Paper](https://arxiv.org/abs/2510.00523)
|
| 45 |
+
- [Webpage](https://sony.github.io/virtue/)
|
| 46 |
+
- [Repository](https://github.com/sony/virtue)
|
| 47 |
+
|
| 48 |
+
## How to Use
|
| 49 |
+
```=python
|
| 50 |
+
import os
|
| 51 |
+
import sys
|
| 52 |
+
import torch
|
| 53 |
+
import numpy as np
|
| 54 |
+
import json
|
| 55 |
+
import hydra
|
| 56 |
+
from hydra.core.global_hydra import GlobalHydra
|
| 57 |
+
from PIL import Image
|
| 58 |
+
|
| 59 |
+
# Add parent directory to path for src imports
|
| 60 |
+
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
|
| 61 |
+
|
| 62 |
+
from src.arguments import ModelArguments, DataArguments, TrainingArguments
|
| 63 |
+
from src.model.model import MMEBModel
|
| 64 |
+
from src.model.processor import load_processor, VLM_IMAGE_TOKENS, get_backbone_name, process_vlm_inputs_fns
|
| 65 |
+
from transformers import AutoConfig
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
# Initialize Hydra for SAM2 loading
|
| 69 |
+
if not GlobalHydra().is_initialized():
|
| 70 |
+
hydra.initialize(config_path="./configs", version_base=None)
|
| 71 |
+
|
| 72 |
+
# Determinism
|
| 73 |
+
torch.manual_seed(42)
|
| 74 |
+
torch.cuda.manual_seed_all(42)
|
| 75 |
+
torch.backends.cudnn.deterministic = True
|
| 76 |
+
torch.backends.cudnn.benchmark = False
|
| 77 |
+
np.random.seed(42)
|
| 78 |
+
|
| 79 |
+
model_dir = 'Sony/VIRTUE-2B-SCaR'
|
| 80 |
+
device = 'cuda' if torch.cuda.is_available() else 'cpu'
|
| 81 |
+
|
| 82 |
+
config = AutoConfig.from_pretrained(model_dir, trust_remote_code=True, token=True)
|
| 83 |
+
|
| 84 |
+
# Build arguments directly (no YAML required)
|
| 85 |
+
model_args = ModelArguments(
|
| 86 |
+
model_name=model_dir,
|
| 87 |
+
checkpoint_path=None,
|
| 88 |
+
pooling="last",
|
| 89 |
+
normalize=True,
|
| 90 |
+
lora=False,
|
| 91 |
+
model_backbone='qwen2_vl',
|
| 92 |
+
)
|
| 93 |
+
persisted_sam = config.virtue_sam
|
| 94 |
+
|
| 95 |
+
model_args.sam = True
|
| 96 |
+
model_args.sam_config = {
|
| 97 |
+
"config_path": persisted_sam.get('config_path') if persisted_sam else None,
|
| 98 |
+
"checkpoint": persisted_sam.get('checkpoint') if persisted_sam else None,
|
| 99 |
+
"points_per_side": (persisted_sam.get('points_per_side') if persisted_sam else 16),
|
| 100 |
+
"feature_levels": (persisted_sam.get('feature_levels') if persisted_sam else 3),
|
| 101 |
+
}
|
| 102 |
+
|
| 103 |
+
data_args = DataArguments()
|
| 104 |
+
training_args = TrainingArguments()
|
| 105 |
+
|
| 106 |
+
processor = load_processor(model_args, data_args)
|
| 107 |
+
model = MMEBModel.load(model_args, is_trainable=False, processor=processor)
|
| 108 |
+
model.eval()
|
| 109 |
+
model = model.to(device, dtype=torch.bfloat16)
|
| 110 |
+
|
| 111 |
+
# Get model backbone and image token
|
| 112 |
+
model_backbone = get_backbone_name(hf_config=config)
|
| 113 |
+
image_token = VLM_IMAGE_TOKENS[model_backbone]
|
| 114 |
+
|
| 115 |
+
# Image + Text -> Text
|
| 116 |
+
image_path = '../assets/example.jpg'
|
| 117 |
+
image = Image.open(image_path).convert('RGB')
|
| 118 |
+
|
| 119 |
+
model_inputs = {
|
| 120 |
+
'text': [f"{image_token}\nRepresent the given image with the following question: What is in the image"],
|
| 121 |
+
'images': [image]
|
| 122 |
+
}
|
| 123 |
+
|
| 124 |
+
process_fn = process_vlm_inputs_fns[model_backbone]
|
| 125 |
+
inputs = process_fn(model_inputs, processor=processor, max_length=512)
|
| 126 |
+
device = next(model.parameters()).device
|
| 127 |
+
inputs = {k: v.to(device) if torch.is_tensor(v) else v for k, v in inputs.items()}
|
| 128 |
+
|
| 129 |
+
with torch.no_grad():
|
| 130 |
+
with torch.autocast(enabled=True, dtype=torch.bfloat16, device_type="cuda"):
|
| 131 |
+
qry_output = model(qry=inputs)["qry_reps"]
|
| 132 |
+
|
| 133 |
+
# Candidates for all scenarios
|
| 134 |
+
test_strings = ['A cat', 'A dog', 'A tiger']
|
| 135 |
+
|
| 136 |
+
# Scenario 1: No visual prompts (image only)
|
| 137 |
+
print("\n--- Similarities (no visual prompts) ---")
|
| 138 |
+
for string in test_strings:
|
| 139 |
+
cand_inputs = process_fn({'text': [string], 'images': [None]}, processor=processor)
|
| 140 |
+
cand_inputs = {k: v.to(device) if torch.is_tensor(v) else v for k, v in cand_inputs.items()}
|
| 141 |
+
with torch.no_grad():
|
| 142 |
+
with torch.autocast(enabled=True, dtype=torch.bfloat16, device_type="cuda"):
|
| 143 |
+
tgt_output = model(tgt=cand_inputs)["tgt_reps"]
|
| 144 |
+
sim = model.compute_similarity(qry_output, tgt_output)
|
| 145 |
+
print(f"no-prompt | {string} = {sim}")
|
| 146 |
+
|
| 147 |
+
'''
|
| 148 |
+
--- Similarities (no visual prompts) ---
|
| 149 |
+
no-prompt | A cat = tensor([[0.3030]], device='cuda:0')
|
| 150 |
+
no-prompt | A dog = tensor([[0.2453]], device='cuda:0')
|
| 151 |
+
no-prompt | A tiger = tensor([[0.1714]], device='cuda:0')
|
| 152 |
+
'''
|
| 153 |
+
|
| 154 |
+
# Scenario 2: Point prompts — two examples (left/right)
|
| 155 |
+
print("\n--- Similarities (point prompts) ---")
|
| 156 |
+
sam_size = 1024 # SAM2Transforms output size
|
| 157 |
+
point_examples = [(0.25, 0.5), (0.75, 0.5)]
|
| 158 |
+
for (px, py) in point_examples:
|
| 159 |
+
point_text = f"{image_token}\nFind the caption that best describes the segmented object, considering both local details and global context in the given image.\nReferring object point: ({int(px*image.size[0])}, {int(py*image.size[1])})"
|
| 160 |
+
q_inputs = process_fn({'text': [point_text], 'images': [image]}, processor=processor)
|
| 161 |
+
q_inputs['point'] = [px * sam_size, py * sam_size]
|
| 162 |
+
q_inputs = {k: v.to(device) if torch.is_tensor(v) else v for k, v in q_inputs.items()}
|
| 163 |
+
with torch.no_grad():
|
| 164 |
+
with torch.autocast(enabled=True, dtype=torch.bfloat16, device_type="cuda"):
|
| 165 |
+
point_qry = model(qry=q_inputs)["qry_reps"]
|
| 166 |
+
for string in test_strings:
|
| 167 |
+
cand_inputs = process_fn({'text': [string], 'images': [None]}, processor=processor)
|
| 168 |
+
cand_inputs = {k: v.to(device) if torch.is_tensor(v) else v for k, v in cand_inputs.items()}
|
| 169 |
+
with torch.no_grad():
|
| 170 |
+
with torch.autocast(enabled=True, dtype=torch.bfloat16, device_type="cuda"):
|
| 171 |
+
tgt_output = model(tgt=cand_inputs)["tgt_reps"]
|
| 172 |
+
sim = model.compute_similarity(point_qry, tgt_output)
|
| 173 |
+
print(f"point ({px:.2f},{py:.2f}) | {string} = {sim}")
|
| 174 |
+
|
| 175 |
+
'''
|
| 176 |
+
--- Similarities (point prompts) ---
|
| 177 |
+
point (0.25,0.50) | A cat = tensor([[0.1793]], device='cuda:0')
|
| 178 |
+
point (0.25,0.50) | A dog = tensor([[0.1339]], device='cuda:0')
|
| 179 |
+
point (0.25,0.50) | A tiger = tensor([[0.1314]], device='cuda:0')
|
| 180 |
+
point (0.75,0.50) | A cat = tensor([[0.2232]], device='cuda:0')
|
| 181 |
+
point (0.75,0.50) | A dog = tensor([[0.1742]], device='cuda:0')
|
| 182 |
+
point (0.75,0.50) | A tiger = tensor([[0.1692]], device='cuda:0')
|
| 183 |
+
'''
|
| 184 |
+
|
| 185 |
+
# Scenario 3: BBox prompts — two examples (left/right)
|
| 186 |
+
print("\n--- Similarities (bbox prompts) ---")
|
| 187 |
+
bbox_examples = [
|
| 188 |
+
(0.05, 0.20, 0.45, 0.80), # left
|
| 189 |
+
(0.55, 0.20, 0.95, 0.80), # right
|
| 190 |
+
]
|
| 191 |
+
for (x1, y1, x2, y2) in bbox_examples:
|
| 192 |
+
bbox_text = f"{image_token}\nFind the caption that best describes the object in the bounding box, considering both local details and global context in the given image.\nReferring object bbox: ({int(x1*image.size[0])}, {int(y1*image.size[1])}, {int(x2*image.size[0])}, {int(y2*image.size[1])})"
|
| 193 |
+
q_inputs = process_fn({'text': [bbox_text], 'images': [image]}, processor=processor)
|
| 194 |
+
q_inputs['bbox'] = [x1 * sam_size, y1 * sam_size, x2 * sam_size, y2 * sam_size]
|
| 195 |
+
q_inputs = {k: v.to(device) if torch.is_tensor(v) else v for k, v in q_inputs.items()}
|
| 196 |
+
with torch.no_grad():
|
| 197 |
+
with torch.autocast(enabled=True, dtype=torch.bfloat16, device_type="cuda"):
|
| 198 |
+
bbox_qry = model(qry=q_inputs)["qry_reps"]
|
| 199 |
+
for string in test_strings:
|
| 200 |
+
cand_inputs = process_fn({'text': [string], 'images': [None]}, processor=processor)
|
| 201 |
+
cand_inputs = {k: v.to(device) if torch.is_tensor(v) else v for k, v in cand_inputs.items()}
|
| 202 |
+
with torch.no_grad():
|
| 203 |
+
with torch.autocast(enabled=True, dtype=torch.bfloat16, device_type="cuda"):
|
| 204 |
+
tgt_output = model(tgt=cand_inputs)["tgt_reps"]
|
| 205 |
+
sim = model.compute_similarity(bbox_qry, tgt_output)
|
| 206 |
+
print(f"bbox ({x1:.2f},{y1:.2f},{x2:.2f},{y2:.2f}) | {string} = {sim}")
|
| 207 |
+
|
| 208 |
+
'''
|
| 209 |
+
--- Similarities (bbox prompts) ---
|
| 210 |
+
bbox (0.05,0.20,0.45,0.80) | A cat = tensor([[0.2100]], device='cuda:0')
|
| 211 |
+
bbox (0.05,0.20,0.45,0.80) | A dog = tensor([[0.1512]], device='cuda:0')
|
| 212 |
+
bbox (0.05,0.20,0.45,0.80) | A tiger = tensor([[0.1719]], device='cuda:0')
|
| 213 |
+
bbox (0.55,0.20,0.95,0.80) | A cat = tensor([[0.1583]], device='cuda:0')
|
| 214 |
+
bbox (0.55,0.20,0.95,0.80) | A dog = tensor([[0.1953]], device='cuda:0')
|
| 215 |
+
bbox (0.55,0.20,0.95,0.80) | A tiger = tensor([[0.1225]], device='cuda:0')
|
| 216 |
+
'''
|
| 217 |
+
```
|
| 218 |
+
|
| 219 |
+
## Ethical Considerations
|
| 220 |
+
_Note: This section is mainly taken from the [AKI](https://huggingface.co/Sony/AKI-4B-phi-3.5-mini) models_.
|
| 221 |
+
|
| 222 |
+
This release is for research purposes only in support of an academic paper. Our models, datasets, and code are not specifically designed or evaluated for all downstream purposes. We strongly recommend users evaluate and address potential concerns related to accuracy, safety, and fairness before deploying this model. We encourage users to consider the common limitations of AI, comply with applicable laws, and leverage best practices when selecting use cases, particularly for high-risk scenarios where errors or misuse could significantly impact people’s lives, rights, or safety.
|
| 223 |
+
|
| 224 |
+
|
| 225 |
+
## Citation
|
| 226 |
+
```
|
| 227 |
+
@article{wangICLR2026virtue,
|
| 228 |
+
author = {Wei-Yao Wang and
|
| 229 |
+
Kazuya Tateishi and
|
| 230 |
+
Qiyu Wu and
|
| 231 |
+
Shusuke Takahashi and
|
| 232 |
+
Yuki Mitsufuji},
|
| 233 |
+
title = {VIRTUE: Visual-Interactive Text-Image Universal Embedder},
|
| 234 |
+
journal = {arXiv preprint arXiv:2510.00523},
|
| 235 |
+
year = {2025}
|
| 236 |
+
}
|
| 237 |
+
```
|
config.json
ADDED
|
@@ -0,0 +1,104 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen2VLForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"attention_dropout": 0.0,
|
| 6 |
+
"bos_token_id": 151643,
|
| 7 |
+
"eos_token_id": 151645,
|
| 8 |
+
"hidden_act": "silu",
|
| 9 |
+
"hidden_size": 3584,
|
| 10 |
+
"image_token_id": 151655,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 18944,
|
| 13 |
+
"max_position_embeddings": 32768,
|
| 14 |
+
"max_window_layers": 28,
|
| 15 |
+
"model_type": "qwen2_vl",
|
| 16 |
+
"num_attention_heads": 28,
|
| 17 |
+
"num_hidden_layers": 28,
|
| 18 |
+
"num_key_value_heads": 4,
|
| 19 |
+
"rms_norm_eps": 1e-06,
|
| 20 |
+
"rope_scaling": {
|
| 21 |
+
"mrope_section": [
|
| 22 |
+
16,
|
| 23 |
+
24,
|
| 24 |
+
24
|
| 25 |
+
],
|
| 26 |
+
"rope_type": "default",
|
| 27 |
+
"type": "default"
|
| 28 |
+
},
|
| 29 |
+
"rope_theta": 1000000.0,
|
| 30 |
+
"sliding_window": 32768,
|
| 31 |
+
"text_config": {
|
| 32 |
+
"architectures": [
|
| 33 |
+
"Qwen2VLForConditionalGeneration"
|
| 34 |
+
],
|
| 35 |
+
"attention_dropout": 0.0,
|
| 36 |
+
"bos_token_id": 151643,
|
| 37 |
+
"eos_token_id": 151645,
|
| 38 |
+
"hidden_act": "silu",
|
| 39 |
+
"hidden_size": 3584,
|
| 40 |
+
"image_token_id": null,
|
| 41 |
+
"initializer_range": 0.02,
|
| 42 |
+
"intermediate_size": 18944,
|
| 43 |
+
"max_position_embeddings": 32768,
|
| 44 |
+
"max_window_layers": 28,
|
| 45 |
+
"model_type": "qwen2_vl_text",
|
| 46 |
+
"num_attention_heads": 28,
|
| 47 |
+
"num_hidden_layers": 28,
|
| 48 |
+
"num_key_value_heads": 4,
|
| 49 |
+
"rms_norm_eps": 1e-06,
|
| 50 |
+
"rope_scaling": {
|
| 51 |
+
"mrope_section": [
|
| 52 |
+
16,
|
| 53 |
+
24,
|
| 54 |
+
24
|
| 55 |
+
],
|
| 56 |
+
"rope_type": "default",
|
| 57 |
+
"type": "default"
|
| 58 |
+
},
|
| 59 |
+
"rope_theta": 1000000.0,
|
| 60 |
+
"sliding_window": 32768,
|
| 61 |
+
"torch_dtype": "bfloat16",
|
| 62 |
+
"use_cache": true,
|
| 63 |
+
"use_sliding_window": false,
|
| 64 |
+
"video_token_id": null,
|
| 65 |
+
"vision_end_token_id": 151653,
|
| 66 |
+
"vision_start_token_id": 151652,
|
| 67 |
+
"vision_token_id": 151654,
|
| 68 |
+
"vocab_size": 152064
|
| 69 |
+
},
|
| 70 |
+
"tie_word_embeddings": false,
|
| 71 |
+
"torch_dtype": "bfloat16",
|
| 72 |
+
"transformers_version": "4.52.3",
|
| 73 |
+
"use_cache": true,
|
| 74 |
+
"use_sliding_window": false,
|
| 75 |
+
"video_token_id": 151656,
|
| 76 |
+
"vision_config": {
|
| 77 |
+
"depth": 32,
|
| 78 |
+
"embed_dim": 1280,
|
| 79 |
+
"hidden_act": "quick_gelu",
|
| 80 |
+
"hidden_size": 3584,
|
| 81 |
+
"in_channels": 3,
|
| 82 |
+
"in_chans": 3,
|
| 83 |
+
"initializer_range": 0.02,
|
| 84 |
+
"mlp_ratio": 4,
|
| 85 |
+
"model_type": "qwen2_vl",
|
| 86 |
+
"num_heads": 16,
|
| 87 |
+
"patch_size": 14,
|
| 88 |
+
"spatial_merge_size": 2,
|
| 89 |
+
"spatial_patch_size": 14,
|
| 90 |
+
"temporal_patch_size": 2,
|
| 91 |
+
"torch_dtype": "bfloat16"
|
| 92 |
+
},
|
| 93 |
+
"vision_end_token_id": 151653,
|
| 94 |
+
"vision_start_token_id": 151652,
|
| 95 |
+
"vision_token_id": 151654,
|
| 96 |
+
"vocab_size": 152064,
|
| 97 |
+
"virtue_sam": {
|
| 98 |
+
"enabled": true,
|
| 99 |
+
"config_path": "./sam2.1/sam2.1_hiera_b+.yaml",
|
| 100 |
+
"checkpoint": "./sam2_checkpoints/sam2.1_hiera_base_plus.pt",
|
| 101 |
+
"points_per_side": 3,
|
| 102 |
+
"feature_levels": 2
|
| 103 |
+
}
|
| 104 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"temperature": 0.01,
|
| 10 |
+
"top_k": 1,
|
| 11 |
+
"top_p": 0.001,
|
| 12 |
+
"transformers_version": "4.52.3"
|
| 13 |
+
}
|
images/MMEB-results-with-SCaR.png
ADDED
|
Git LFS Details
|
images/MMEB-results.png
ADDED
|
Git LFS Details
|
images/SCaR-results.png
ADDED
|
Git LFS Details
|
images/VIRTUE-framework.jpg
ADDED
|
Git LFS Details
|
images/example.jpg
ADDED
|
Git LFS Details
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": true,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.48145466,
|
| 8 |
+
0.4578275,
|
| 9 |
+
0.40821073
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "Qwen2VLImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.26862954,
|
| 14 |
+
0.26130258,
|
| 15 |
+
0.27577711
|
| 16 |
+
],
|
| 17 |
+
"max_pixels": 1003520,
|
| 18 |
+
"merge_size": 2,
|
| 19 |
+
"min_pixels": 3136,
|
| 20 |
+
"patch_size": 14,
|
| 21 |
+
"processor_class": "Qwen2VLProcessor",
|
| 22 |
+
"resample": 3,
|
| 23 |
+
"rescale_factor": 0.00392156862745098,
|
| 24 |
+
"size": {
|
| 25 |
+
"longest_edge": 1003520,
|
| 26 |
+
"shortest_edge": 3136
|
| 27 |
+
},
|
| 28 |
+
"temporal_patch_size": 2
|
| 29 |
+
}
|
pytorch_model-00001-of-00004.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f41c343aca31558725a37353af8bea47e86e1d1e4be03b56c4f7ec897a4cf955
|
| 3 |
+
size 4966769566
|
pytorch_model-00002-of-00004.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6c4beacbb9d19efcafdd1d000350e32d206a934401f90f76d4fa5e29807f09e3
|
| 3 |
+
size 4991527199
|
pytorch_model-00003-of-00004.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:02212be6df376a6df7960a67ca384183ace0ff59b698cd7ce7459f664249e1fe
|
| 3 |
+
size 4932780264
|
pytorch_model-00004-of-00004.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ac0aa08b37e146e9619c8e8f124b1e0a1169ff2ad829e811022e2b252bac446
|
| 3 |
+
size 1691928988
|
pytorch_model.bin.index.json
ADDED
|
@@ -0,0 +1,737 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_size": 16582751232
|
| 4 |
+
},
|
| 5 |
+
"weight_map": {
|
| 6 |
+
"lm_head.weight": "pytorch_model-00004-of-00004.bin",
|
| 7 |
+
"model.embed_tokens.weight": "pytorch_model-00001-of-00004.bin",
|
| 8 |
+
"model.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00004.bin",
|
| 9 |
+
"model.layers.0.mlp.down_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 10 |
+
"model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 11 |
+
"model.layers.0.mlp.up_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 12 |
+
"model.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00004.bin",
|
| 13 |
+
"model.layers.0.self_attn.k_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 14 |
+
"model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 15 |
+
"model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 16 |
+
"model.layers.0.self_attn.q_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 17 |
+
"model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 18 |
+
"model.layers.0.self_attn.v_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 19 |
+
"model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 20 |
+
"model.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00004.bin",
|
| 21 |
+
"model.layers.1.mlp.down_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 22 |
+
"model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 23 |
+
"model.layers.1.mlp.up_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 24 |
+
"model.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00004.bin",
|
| 25 |
+
"model.layers.1.self_attn.k_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 26 |
+
"model.layers.1.self_attn.k_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 27 |
+
"model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 28 |
+
"model.layers.1.self_attn.q_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 29 |
+
"model.layers.1.self_attn.q_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 30 |
+
"model.layers.1.self_attn.v_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 31 |
+
"model.layers.1.self_attn.v_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 32 |
+
"model.layers.10.input_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 33 |
+
"model.layers.10.mlp.down_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 34 |
+
"model.layers.10.mlp.gate_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 35 |
+
"model.layers.10.mlp.up_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 36 |
+
"model.layers.10.post_attention_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 37 |
+
"model.layers.10.self_attn.k_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 38 |
+
"model.layers.10.self_attn.k_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 39 |
+
"model.layers.10.self_attn.o_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 40 |
+
"model.layers.10.self_attn.q_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 41 |
+
"model.layers.10.self_attn.q_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 42 |
+
"model.layers.10.self_attn.v_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 43 |
+
"model.layers.10.self_attn.v_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 44 |
+
"model.layers.11.input_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 45 |
+
"model.layers.11.mlp.down_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 46 |
+
"model.layers.11.mlp.gate_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 47 |
+
"model.layers.11.mlp.up_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 48 |
+
"model.layers.11.post_attention_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 49 |
+
"model.layers.11.self_attn.k_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 50 |
+
"model.layers.11.self_attn.k_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 51 |
+
"model.layers.11.self_attn.o_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 52 |
+
"model.layers.11.self_attn.q_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 53 |
+
"model.layers.11.self_attn.q_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 54 |
+
"model.layers.11.self_attn.v_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 55 |
+
"model.layers.11.self_attn.v_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 56 |
+
"model.layers.12.input_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 57 |
+
"model.layers.12.mlp.down_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 58 |
+
"model.layers.12.mlp.gate_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 59 |
+
"model.layers.12.mlp.up_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 60 |
+
"model.layers.12.post_attention_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 61 |
+
"model.layers.12.self_attn.k_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 62 |
+
"model.layers.12.self_attn.k_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 63 |
+
"model.layers.12.self_attn.o_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 64 |
+
"model.layers.12.self_attn.q_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 65 |
+
"model.layers.12.self_attn.q_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 66 |
+
"model.layers.12.self_attn.v_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 67 |
+
"model.layers.12.self_attn.v_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 68 |
+
"model.layers.13.input_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 69 |
+
"model.layers.13.mlp.down_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 70 |
+
"model.layers.13.mlp.gate_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 71 |
+
"model.layers.13.mlp.up_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 72 |
+
"model.layers.13.post_attention_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 73 |
+
"model.layers.13.self_attn.k_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 74 |
+
"model.layers.13.self_attn.k_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 75 |
+
"model.layers.13.self_attn.o_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 76 |
+
"model.layers.13.self_attn.q_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 77 |
+
"model.layers.13.self_attn.q_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 78 |
+
"model.layers.13.self_attn.v_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 79 |
+
"model.layers.13.self_attn.v_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 80 |
+
"model.layers.14.input_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 81 |
+
"model.layers.14.mlp.down_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 82 |
+
"model.layers.14.mlp.gate_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 83 |
+
"model.layers.14.mlp.up_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 84 |
+
"model.layers.14.post_attention_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 85 |
+
"model.layers.14.self_attn.k_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 86 |
+
"model.layers.14.self_attn.k_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 87 |
+
"model.layers.14.self_attn.o_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 88 |
+
"model.layers.14.self_attn.q_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 89 |
+
"model.layers.14.self_attn.q_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 90 |
+
"model.layers.14.self_attn.v_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 91 |
+
"model.layers.14.self_attn.v_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 92 |
+
"model.layers.15.input_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 93 |
+
"model.layers.15.mlp.down_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 94 |
+
"model.layers.15.mlp.gate_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 95 |
+
"model.layers.15.mlp.up_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 96 |
+
"model.layers.15.post_attention_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 97 |
+
"model.layers.15.self_attn.k_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 98 |
+
"model.layers.15.self_attn.k_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 99 |
+
"model.layers.15.self_attn.o_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 100 |
+
"model.layers.15.self_attn.q_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 101 |
+
"model.layers.15.self_attn.q_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 102 |
+
"model.layers.15.self_attn.v_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 103 |
+
"model.layers.15.self_attn.v_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 104 |
+
"model.layers.16.input_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 105 |
+
"model.layers.16.mlp.down_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 106 |
+
"model.layers.16.mlp.gate_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 107 |
+
"model.layers.16.mlp.up_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 108 |
+
"model.layers.16.post_attention_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 109 |
+
"model.layers.16.self_attn.k_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 110 |
+
"model.layers.16.self_attn.k_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 111 |
+
"model.layers.16.self_attn.o_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 112 |
+
"model.layers.16.self_attn.q_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 113 |
+
"model.layers.16.self_attn.q_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 114 |
+
"model.layers.16.self_attn.v_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 115 |
+
"model.layers.16.self_attn.v_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 116 |
+
"model.layers.17.input_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 117 |
+
"model.layers.17.mlp.down_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 118 |
+
"model.layers.17.mlp.gate_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 119 |
+
"model.layers.17.mlp.up_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 120 |
+
"model.layers.17.post_attention_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 121 |
+
"model.layers.17.self_attn.k_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 122 |
+
"model.layers.17.self_attn.k_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 123 |
+
"model.layers.17.self_attn.o_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 124 |
+
"model.layers.17.self_attn.q_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 125 |
+
"model.layers.17.self_attn.q_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 126 |
+
"model.layers.17.self_attn.v_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 127 |
+
"model.layers.17.self_attn.v_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 128 |
+
"model.layers.18.input_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 129 |
+
"model.layers.18.mlp.down_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 130 |
+
"model.layers.18.mlp.gate_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 131 |
+
"model.layers.18.mlp.up_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 132 |
+
"model.layers.18.post_attention_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 133 |
+
"model.layers.18.self_attn.k_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 134 |
+
"model.layers.18.self_attn.k_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 135 |
+
"model.layers.18.self_attn.o_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 136 |
+
"model.layers.18.self_attn.q_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 137 |
+
"model.layers.18.self_attn.q_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 138 |
+
"model.layers.18.self_attn.v_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 139 |
+
"model.layers.18.self_attn.v_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 140 |
+
"model.layers.19.input_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 141 |
+
"model.layers.19.mlp.down_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 142 |
+
"model.layers.19.mlp.gate_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 143 |
+
"model.layers.19.mlp.up_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 144 |
+
"model.layers.19.post_attention_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 145 |
+
"model.layers.19.self_attn.k_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 146 |
+
"model.layers.19.self_attn.k_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 147 |
+
"model.layers.19.self_attn.o_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 148 |
+
"model.layers.19.self_attn.q_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 149 |
+
"model.layers.19.self_attn.q_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 150 |
+
"model.layers.19.self_attn.v_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 151 |
+
"model.layers.19.self_attn.v_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 152 |
+
"model.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00004.bin",
|
| 153 |
+
"model.layers.2.mlp.down_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 154 |
+
"model.layers.2.mlp.gate_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 155 |
+
"model.layers.2.mlp.up_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 156 |
+
"model.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00004.bin",
|
| 157 |
+
"model.layers.2.self_attn.k_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 158 |
+
"model.layers.2.self_attn.k_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 159 |
+
"model.layers.2.self_attn.o_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 160 |
+
"model.layers.2.self_attn.q_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 161 |
+
"model.layers.2.self_attn.q_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 162 |
+
"model.layers.2.self_attn.v_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 163 |
+
"model.layers.2.self_attn.v_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 164 |
+
"model.layers.20.input_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 165 |
+
"model.layers.20.mlp.down_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 166 |
+
"model.layers.20.mlp.gate_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 167 |
+
"model.layers.20.mlp.up_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 168 |
+
"model.layers.20.post_attention_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 169 |
+
"model.layers.20.self_attn.k_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 170 |
+
"model.layers.20.self_attn.k_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 171 |
+
"model.layers.20.self_attn.o_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 172 |
+
"model.layers.20.self_attn.q_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 173 |
+
"model.layers.20.self_attn.q_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 174 |
+
"model.layers.20.self_attn.v_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 175 |
+
"model.layers.20.self_attn.v_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 176 |
+
"model.layers.21.input_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 177 |
+
"model.layers.21.mlp.down_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 178 |
+
"model.layers.21.mlp.gate_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 179 |
+
"model.layers.21.mlp.up_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 180 |
+
"model.layers.21.post_attention_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 181 |
+
"model.layers.21.self_attn.k_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 182 |
+
"model.layers.21.self_attn.k_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 183 |
+
"model.layers.21.self_attn.o_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 184 |
+
"model.layers.21.self_attn.q_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 185 |
+
"model.layers.21.self_attn.q_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 186 |
+
"model.layers.21.self_attn.v_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 187 |
+
"model.layers.21.self_attn.v_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 188 |
+
"model.layers.22.input_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 189 |
+
"model.layers.22.mlp.down_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 190 |
+
"model.layers.22.mlp.gate_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 191 |
+
"model.layers.22.mlp.up_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 192 |
+
"model.layers.22.post_attention_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 193 |
+
"model.layers.22.self_attn.k_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 194 |
+
"model.layers.22.self_attn.k_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 195 |
+
"model.layers.22.self_attn.o_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 196 |
+
"model.layers.22.self_attn.q_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 197 |
+
"model.layers.22.self_attn.q_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 198 |
+
"model.layers.22.self_attn.v_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 199 |
+
"model.layers.22.self_attn.v_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 200 |
+
"model.layers.23.input_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 201 |
+
"model.layers.23.mlp.down_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 202 |
+
"model.layers.23.mlp.gate_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 203 |
+
"model.layers.23.mlp.up_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 204 |
+
"model.layers.23.post_attention_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 205 |
+
"model.layers.23.self_attn.k_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 206 |
+
"model.layers.23.self_attn.k_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 207 |
+
"model.layers.23.self_attn.o_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 208 |
+
"model.layers.23.self_attn.q_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 209 |
+
"model.layers.23.self_attn.q_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 210 |
+
"model.layers.23.self_attn.v_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 211 |
+
"model.layers.23.self_attn.v_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 212 |
+
"model.layers.24.input_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 213 |
+
"model.layers.24.mlp.down_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 214 |
+
"model.layers.24.mlp.gate_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 215 |
+
"model.layers.24.mlp.up_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 216 |
+
"model.layers.24.post_attention_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 217 |
+
"model.layers.24.self_attn.k_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 218 |
+
"model.layers.24.self_attn.k_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 219 |
+
"model.layers.24.self_attn.o_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 220 |
+
"model.layers.24.self_attn.q_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 221 |
+
"model.layers.24.self_attn.q_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 222 |
+
"model.layers.24.self_attn.v_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 223 |
+
"model.layers.24.self_attn.v_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 224 |
+
"model.layers.25.input_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 225 |
+
"model.layers.25.mlp.down_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 226 |
+
"model.layers.25.mlp.gate_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 227 |
+
"model.layers.25.mlp.up_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 228 |
+
"model.layers.25.post_attention_layernorm.weight": "pytorch_model-00003-of-00004.bin",
|
| 229 |
+
"model.layers.25.self_attn.k_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 230 |
+
"model.layers.25.self_attn.k_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 231 |
+
"model.layers.25.self_attn.o_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 232 |
+
"model.layers.25.self_attn.q_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 233 |
+
"model.layers.25.self_attn.q_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 234 |
+
"model.layers.25.self_attn.v_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 235 |
+
"model.layers.25.self_attn.v_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 236 |
+
"model.layers.26.input_layernorm.weight": "pytorch_model-00004-of-00004.bin",
|
| 237 |
+
"model.layers.26.mlp.down_proj.weight": "pytorch_model-00004-of-00004.bin",
|
| 238 |
+
"model.layers.26.mlp.gate_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 239 |
+
"model.layers.26.mlp.up_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 240 |
+
"model.layers.26.post_attention_layernorm.weight": "pytorch_model-00004-of-00004.bin",
|
| 241 |
+
"model.layers.26.self_attn.k_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 242 |
+
"model.layers.26.self_attn.k_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 243 |
+
"model.layers.26.self_attn.o_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 244 |
+
"model.layers.26.self_attn.q_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 245 |
+
"model.layers.26.self_attn.q_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 246 |
+
"model.layers.26.self_attn.v_proj.bias": "pytorch_model-00003-of-00004.bin",
|
| 247 |
+
"model.layers.26.self_attn.v_proj.weight": "pytorch_model-00003-of-00004.bin",
|
| 248 |
+
"model.layers.27.input_layernorm.weight": "pytorch_model-00004-of-00004.bin",
|
| 249 |
+
"model.layers.27.mlp.down_proj.weight": "pytorch_model-00004-of-00004.bin",
|
| 250 |
+
"model.layers.27.mlp.gate_proj.weight": "pytorch_model-00004-of-00004.bin",
|
| 251 |
+
"model.layers.27.mlp.up_proj.weight": "pytorch_model-00004-of-00004.bin",
|
| 252 |
+
"model.layers.27.post_attention_layernorm.weight": "pytorch_model-00004-of-00004.bin",
|
| 253 |
+
"model.layers.27.self_attn.k_proj.bias": "pytorch_model-00004-of-00004.bin",
|
| 254 |
+
"model.layers.27.self_attn.k_proj.weight": "pytorch_model-00004-of-00004.bin",
|
| 255 |
+
"model.layers.27.self_attn.o_proj.weight": "pytorch_model-00004-of-00004.bin",
|
| 256 |
+
"model.layers.27.self_attn.q_proj.bias": "pytorch_model-00004-of-00004.bin",
|
| 257 |
+
"model.layers.27.self_attn.q_proj.weight": "pytorch_model-00004-of-00004.bin",
|
| 258 |
+
"model.layers.27.self_attn.v_proj.bias": "pytorch_model-00004-of-00004.bin",
|
| 259 |
+
"model.layers.27.self_attn.v_proj.weight": "pytorch_model-00004-of-00004.bin",
|
| 260 |
+
"model.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00004.bin",
|
| 261 |
+
"model.layers.3.mlp.down_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 262 |
+
"model.layers.3.mlp.gate_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 263 |
+
"model.layers.3.mlp.up_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 264 |
+
"model.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00004.bin",
|
| 265 |
+
"model.layers.3.self_attn.k_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 266 |
+
"model.layers.3.self_attn.k_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 267 |
+
"model.layers.3.self_attn.o_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 268 |
+
"model.layers.3.self_attn.q_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 269 |
+
"model.layers.3.self_attn.q_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 270 |
+
"model.layers.3.self_attn.v_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 271 |
+
"model.layers.3.self_attn.v_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 272 |
+
"model.layers.4.input_layernorm.weight": "pytorch_model-00001-of-00004.bin",
|
| 273 |
+
"model.layers.4.mlp.down_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 274 |
+
"model.layers.4.mlp.gate_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 275 |
+
"model.layers.4.mlp.up_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 276 |
+
"model.layers.4.post_attention_layernorm.weight": "pytorch_model-00001-of-00004.bin",
|
| 277 |
+
"model.layers.4.self_attn.k_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 278 |
+
"model.layers.4.self_attn.k_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 279 |
+
"model.layers.4.self_attn.o_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 280 |
+
"model.layers.4.self_attn.q_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 281 |
+
"model.layers.4.self_attn.q_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 282 |
+
"model.layers.4.self_attn.v_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 283 |
+
"model.layers.4.self_attn.v_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 284 |
+
"model.layers.5.input_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 285 |
+
"model.layers.5.mlp.down_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 286 |
+
"model.layers.5.mlp.gate_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 287 |
+
"model.layers.5.mlp.up_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 288 |
+
"model.layers.5.post_attention_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 289 |
+
"model.layers.5.self_attn.k_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 290 |
+
"model.layers.5.self_attn.k_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 291 |
+
"model.layers.5.self_attn.o_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 292 |
+
"model.layers.5.self_attn.q_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 293 |
+
"model.layers.5.self_attn.q_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 294 |
+
"model.layers.5.self_attn.v_proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 295 |
+
"model.layers.5.self_attn.v_proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 296 |
+
"model.layers.6.input_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 297 |
+
"model.layers.6.mlp.down_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 298 |
+
"model.layers.6.mlp.gate_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 299 |
+
"model.layers.6.mlp.up_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 300 |
+
"model.layers.6.post_attention_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 301 |
+
"model.layers.6.self_attn.k_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 302 |
+
"model.layers.6.self_attn.k_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 303 |
+
"model.layers.6.self_attn.o_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 304 |
+
"model.layers.6.self_attn.q_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 305 |
+
"model.layers.6.self_attn.q_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 306 |
+
"model.layers.6.self_attn.v_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 307 |
+
"model.layers.6.self_attn.v_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 308 |
+
"model.layers.7.input_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 309 |
+
"model.layers.7.mlp.down_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 310 |
+
"model.layers.7.mlp.gate_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 311 |
+
"model.layers.7.mlp.up_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 312 |
+
"model.layers.7.post_attention_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 313 |
+
"model.layers.7.self_attn.k_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 314 |
+
"model.layers.7.self_attn.k_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 315 |
+
"model.layers.7.self_attn.o_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 316 |
+
"model.layers.7.self_attn.q_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 317 |
+
"model.layers.7.self_attn.q_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 318 |
+
"model.layers.7.self_attn.v_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 319 |
+
"model.layers.7.self_attn.v_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 320 |
+
"model.layers.8.input_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 321 |
+
"model.layers.8.mlp.down_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 322 |
+
"model.layers.8.mlp.gate_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 323 |
+
"model.layers.8.mlp.up_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 324 |
+
"model.layers.8.post_attention_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 325 |
+
"model.layers.8.self_attn.k_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 326 |
+
"model.layers.8.self_attn.k_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 327 |
+
"model.layers.8.self_attn.o_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 328 |
+
"model.layers.8.self_attn.q_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 329 |
+
"model.layers.8.self_attn.q_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 330 |
+
"model.layers.8.self_attn.v_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 331 |
+
"model.layers.8.self_attn.v_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 332 |
+
"model.layers.9.input_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 333 |
+
"model.layers.9.mlp.down_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 334 |
+
"model.layers.9.mlp.gate_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 335 |
+
"model.layers.9.mlp.up_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 336 |
+
"model.layers.9.post_attention_layernorm.weight": "pytorch_model-00002-of-00004.bin",
|
| 337 |
+
"model.layers.9.self_attn.k_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 338 |
+
"model.layers.9.self_attn.k_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 339 |
+
"model.layers.9.self_attn.o_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 340 |
+
"model.layers.9.self_attn.q_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 341 |
+
"model.layers.9.self_attn.q_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 342 |
+
"model.layers.9.self_attn.v_proj.bias": "pytorch_model-00002-of-00004.bin",
|
| 343 |
+
"model.layers.9.self_attn.v_proj.weight": "pytorch_model-00002-of-00004.bin",
|
| 344 |
+
"model.norm.weight": "pytorch_model-00004-of-00004.bin",
|
| 345 |
+
"visual.blocks.0.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 346 |
+
"visual.blocks.0.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 347 |
+
"visual.blocks.0.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 348 |
+
"visual.blocks.0.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 349 |
+
"visual.blocks.0.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 350 |
+
"visual.blocks.0.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 351 |
+
"visual.blocks.0.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 352 |
+
"visual.blocks.0.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 353 |
+
"visual.blocks.0.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 354 |
+
"visual.blocks.0.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 355 |
+
"visual.blocks.0.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 356 |
+
"visual.blocks.0.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 357 |
+
"visual.blocks.1.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 358 |
+
"visual.blocks.1.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 359 |
+
"visual.blocks.1.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 360 |
+
"visual.blocks.1.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 361 |
+
"visual.blocks.1.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 362 |
+
"visual.blocks.1.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 363 |
+
"visual.blocks.1.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 364 |
+
"visual.blocks.1.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 365 |
+
"visual.blocks.1.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 366 |
+
"visual.blocks.1.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 367 |
+
"visual.blocks.1.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 368 |
+
"visual.blocks.1.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 369 |
+
"visual.blocks.10.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 370 |
+
"visual.blocks.10.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 371 |
+
"visual.blocks.10.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 372 |
+
"visual.blocks.10.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 373 |
+
"visual.blocks.10.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 374 |
+
"visual.blocks.10.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 375 |
+
"visual.blocks.10.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 376 |
+
"visual.blocks.10.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 377 |
+
"visual.blocks.10.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 378 |
+
"visual.blocks.10.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 379 |
+
"visual.blocks.10.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 380 |
+
"visual.blocks.10.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 381 |
+
"visual.blocks.11.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 382 |
+
"visual.blocks.11.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 383 |
+
"visual.blocks.11.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 384 |
+
"visual.blocks.11.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 385 |
+
"visual.blocks.11.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 386 |
+
"visual.blocks.11.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 387 |
+
"visual.blocks.11.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 388 |
+
"visual.blocks.11.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 389 |
+
"visual.blocks.11.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 390 |
+
"visual.blocks.11.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 391 |
+
"visual.blocks.11.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 392 |
+
"visual.blocks.11.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 393 |
+
"visual.blocks.12.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 394 |
+
"visual.blocks.12.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 395 |
+
"visual.blocks.12.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 396 |
+
"visual.blocks.12.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 397 |
+
"visual.blocks.12.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 398 |
+
"visual.blocks.12.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 399 |
+
"visual.blocks.12.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 400 |
+
"visual.blocks.12.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 401 |
+
"visual.blocks.12.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 402 |
+
"visual.blocks.12.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 403 |
+
"visual.blocks.12.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 404 |
+
"visual.blocks.12.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 405 |
+
"visual.blocks.13.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 406 |
+
"visual.blocks.13.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 407 |
+
"visual.blocks.13.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 408 |
+
"visual.blocks.13.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 409 |
+
"visual.blocks.13.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 410 |
+
"visual.blocks.13.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 411 |
+
"visual.blocks.13.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 412 |
+
"visual.blocks.13.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 413 |
+
"visual.blocks.13.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 414 |
+
"visual.blocks.13.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 415 |
+
"visual.blocks.13.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 416 |
+
"visual.blocks.13.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 417 |
+
"visual.blocks.14.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 418 |
+
"visual.blocks.14.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 419 |
+
"visual.blocks.14.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 420 |
+
"visual.blocks.14.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 421 |
+
"visual.blocks.14.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 422 |
+
"visual.blocks.14.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 423 |
+
"visual.blocks.14.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 424 |
+
"visual.blocks.14.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 425 |
+
"visual.blocks.14.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 426 |
+
"visual.blocks.14.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 427 |
+
"visual.blocks.14.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 428 |
+
"visual.blocks.14.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 429 |
+
"visual.blocks.15.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 430 |
+
"visual.blocks.15.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 431 |
+
"visual.blocks.15.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 432 |
+
"visual.blocks.15.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 433 |
+
"visual.blocks.15.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 434 |
+
"visual.blocks.15.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 435 |
+
"visual.blocks.15.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 436 |
+
"visual.blocks.15.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 437 |
+
"visual.blocks.15.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 438 |
+
"visual.blocks.15.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 439 |
+
"visual.blocks.15.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 440 |
+
"visual.blocks.15.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 441 |
+
"visual.blocks.16.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 442 |
+
"visual.blocks.16.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 443 |
+
"visual.blocks.16.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 444 |
+
"visual.blocks.16.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 445 |
+
"visual.blocks.16.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 446 |
+
"visual.blocks.16.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 447 |
+
"visual.blocks.16.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 448 |
+
"visual.blocks.16.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 449 |
+
"visual.blocks.16.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 450 |
+
"visual.blocks.16.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 451 |
+
"visual.blocks.16.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 452 |
+
"visual.blocks.16.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 453 |
+
"visual.blocks.17.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 454 |
+
"visual.blocks.17.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 455 |
+
"visual.blocks.17.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 456 |
+
"visual.blocks.17.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 457 |
+
"visual.blocks.17.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 458 |
+
"visual.blocks.17.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 459 |
+
"visual.blocks.17.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 460 |
+
"visual.blocks.17.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 461 |
+
"visual.blocks.17.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 462 |
+
"visual.blocks.17.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 463 |
+
"visual.blocks.17.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 464 |
+
"visual.blocks.17.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 465 |
+
"visual.blocks.18.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 466 |
+
"visual.blocks.18.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 467 |
+
"visual.blocks.18.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 468 |
+
"visual.blocks.18.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 469 |
+
"visual.blocks.18.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 470 |
+
"visual.blocks.18.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 471 |
+
"visual.blocks.18.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 472 |
+
"visual.blocks.18.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 473 |
+
"visual.blocks.18.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 474 |
+
"visual.blocks.18.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 475 |
+
"visual.blocks.18.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 476 |
+
"visual.blocks.18.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 477 |
+
"visual.blocks.19.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 478 |
+
"visual.blocks.19.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 479 |
+
"visual.blocks.19.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 480 |
+
"visual.blocks.19.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 481 |
+
"visual.blocks.19.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 482 |
+
"visual.blocks.19.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 483 |
+
"visual.blocks.19.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 484 |
+
"visual.blocks.19.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 485 |
+
"visual.blocks.19.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 486 |
+
"visual.blocks.19.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 487 |
+
"visual.blocks.19.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 488 |
+
"visual.blocks.19.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 489 |
+
"visual.blocks.2.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 490 |
+
"visual.blocks.2.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 491 |
+
"visual.blocks.2.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 492 |
+
"visual.blocks.2.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 493 |
+
"visual.blocks.2.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 494 |
+
"visual.blocks.2.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 495 |
+
"visual.blocks.2.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 496 |
+
"visual.blocks.2.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 497 |
+
"visual.blocks.2.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 498 |
+
"visual.blocks.2.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 499 |
+
"visual.blocks.2.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 500 |
+
"visual.blocks.2.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 501 |
+
"visual.blocks.20.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 502 |
+
"visual.blocks.20.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 503 |
+
"visual.blocks.20.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 504 |
+
"visual.blocks.20.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 505 |
+
"visual.blocks.20.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 506 |
+
"visual.blocks.20.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 507 |
+
"visual.blocks.20.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 508 |
+
"visual.blocks.20.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 509 |
+
"visual.blocks.20.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 510 |
+
"visual.blocks.20.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 511 |
+
"visual.blocks.20.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 512 |
+
"visual.blocks.20.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 513 |
+
"visual.blocks.21.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 514 |
+
"visual.blocks.21.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 515 |
+
"visual.blocks.21.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 516 |
+
"visual.blocks.21.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 517 |
+
"visual.blocks.21.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 518 |
+
"visual.blocks.21.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 519 |
+
"visual.blocks.21.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 520 |
+
"visual.blocks.21.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 521 |
+
"visual.blocks.21.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 522 |
+
"visual.blocks.21.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 523 |
+
"visual.blocks.21.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 524 |
+
"visual.blocks.21.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 525 |
+
"visual.blocks.22.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 526 |
+
"visual.blocks.22.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 527 |
+
"visual.blocks.22.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 528 |
+
"visual.blocks.22.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 529 |
+
"visual.blocks.22.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 530 |
+
"visual.blocks.22.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 531 |
+
"visual.blocks.22.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 532 |
+
"visual.blocks.22.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 533 |
+
"visual.blocks.22.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 534 |
+
"visual.blocks.22.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 535 |
+
"visual.blocks.22.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 536 |
+
"visual.blocks.22.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 537 |
+
"visual.blocks.23.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 538 |
+
"visual.blocks.23.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 539 |
+
"visual.blocks.23.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 540 |
+
"visual.blocks.23.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 541 |
+
"visual.blocks.23.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 542 |
+
"visual.blocks.23.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 543 |
+
"visual.blocks.23.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 544 |
+
"visual.blocks.23.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 545 |
+
"visual.blocks.23.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 546 |
+
"visual.blocks.23.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 547 |
+
"visual.blocks.23.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 548 |
+
"visual.blocks.23.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 549 |
+
"visual.blocks.24.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 550 |
+
"visual.blocks.24.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 551 |
+
"visual.blocks.24.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 552 |
+
"visual.blocks.24.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 553 |
+
"visual.blocks.24.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 554 |
+
"visual.blocks.24.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 555 |
+
"visual.blocks.24.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 556 |
+
"visual.blocks.24.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 557 |
+
"visual.blocks.24.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 558 |
+
"visual.blocks.24.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 559 |
+
"visual.blocks.24.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 560 |
+
"visual.blocks.24.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 561 |
+
"visual.blocks.25.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 562 |
+
"visual.blocks.25.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 563 |
+
"visual.blocks.25.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 564 |
+
"visual.blocks.25.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 565 |
+
"visual.blocks.25.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 566 |
+
"visual.blocks.25.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 567 |
+
"visual.blocks.25.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 568 |
+
"visual.blocks.25.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 569 |
+
"visual.blocks.25.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 570 |
+
"visual.blocks.25.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 571 |
+
"visual.blocks.25.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 572 |
+
"visual.blocks.25.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 573 |
+
"visual.blocks.26.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 574 |
+
"visual.blocks.26.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 575 |
+
"visual.blocks.26.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 576 |
+
"visual.blocks.26.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 577 |
+
"visual.blocks.26.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 578 |
+
"visual.blocks.26.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 579 |
+
"visual.blocks.26.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 580 |
+
"visual.blocks.26.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 581 |
+
"visual.blocks.26.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 582 |
+
"visual.blocks.26.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 583 |
+
"visual.blocks.26.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 584 |
+
"visual.blocks.26.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 585 |
+
"visual.blocks.27.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 586 |
+
"visual.blocks.27.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 587 |
+
"visual.blocks.27.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 588 |
+
"visual.blocks.27.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 589 |
+
"visual.blocks.27.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 590 |
+
"visual.blocks.27.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 591 |
+
"visual.blocks.27.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 592 |
+
"visual.blocks.27.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 593 |
+
"visual.blocks.27.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 594 |
+
"visual.blocks.27.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 595 |
+
"visual.blocks.27.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 596 |
+
"visual.blocks.27.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 597 |
+
"visual.blocks.28.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 598 |
+
"visual.blocks.28.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 599 |
+
"visual.blocks.28.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 600 |
+
"visual.blocks.28.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 601 |
+
"visual.blocks.28.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 602 |
+
"visual.blocks.28.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 603 |
+
"visual.blocks.28.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 604 |
+
"visual.blocks.28.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 605 |
+
"visual.blocks.28.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 606 |
+
"visual.blocks.28.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 607 |
+
"visual.blocks.28.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 608 |
+
"visual.blocks.28.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 609 |
+
"visual.blocks.29.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 610 |
+
"visual.blocks.29.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 611 |
+
"visual.blocks.29.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 612 |
+
"visual.blocks.29.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 613 |
+
"visual.blocks.29.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 614 |
+
"visual.blocks.29.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 615 |
+
"visual.blocks.29.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 616 |
+
"visual.blocks.29.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 617 |
+
"visual.blocks.29.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 618 |
+
"visual.blocks.29.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 619 |
+
"visual.blocks.29.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 620 |
+
"visual.blocks.29.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 621 |
+
"visual.blocks.3.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 622 |
+
"visual.blocks.3.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 623 |
+
"visual.blocks.3.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 624 |
+
"visual.blocks.3.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 625 |
+
"visual.blocks.3.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 626 |
+
"visual.blocks.3.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 627 |
+
"visual.blocks.3.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 628 |
+
"visual.blocks.3.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 629 |
+
"visual.blocks.3.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 630 |
+
"visual.blocks.3.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 631 |
+
"visual.blocks.3.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 632 |
+
"visual.blocks.3.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 633 |
+
"visual.blocks.30.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 634 |
+
"visual.blocks.30.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 635 |
+
"visual.blocks.30.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 636 |
+
"visual.blocks.30.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 637 |
+
"visual.blocks.30.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 638 |
+
"visual.blocks.30.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 639 |
+
"visual.blocks.30.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 640 |
+
"visual.blocks.30.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 641 |
+
"visual.blocks.30.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 642 |
+
"visual.blocks.30.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 643 |
+
"visual.blocks.30.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 644 |
+
"visual.blocks.30.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 645 |
+
"visual.blocks.31.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 646 |
+
"visual.blocks.31.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 647 |
+
"visual.blocks.31.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 648 |
+
"visual.blocks.31.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 649 |
+
"visual.blocks.31.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 650 |
+
"visual.blocks.31.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 651 |
+
"visual.blocks.31.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 652 |
+
"visual.blocks.31.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 653 |
+
"visual.blocks.31.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 654 |
+
"visual.blocks.31.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 655 |
+
"visual.blocks.31.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 656 |
+
"visual.blocks.31.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 657 |
+
"visual.blocks.4.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 658 |
+
"visual.blocks.4.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 659 |
+
"visual.blocks.4.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 660 |
+
"visual.blocks.4.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 661 |
+
"visual.blocks.4.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 662 |
+
"visual.blocks.4.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 663 |
+
"visual.blocks.4.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 664 |
+
"visual.blocks.4.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 665 |
+
"visual.blocks.4.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 666 |
+
"visual.blocks.4.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 667 |
+
"visual.blocks.4.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 668 |
+
"visual.blocks.4.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 669 |
+
"visual.blocks.5.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 670 |
+
"visual.blocks.5.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 671 |
+
"visual.blocks.5.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 672 |
+
"visual.blocks.5.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 673 |
+
"visual.blocks.5.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 674 |
+
"visual.blocks.5.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 675 |
+
"visual.blocks.5.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 676 |
+
"visual.blocks.5.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 677 |
+
"visual.blocks.5.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 678 |
+
"visual.blocks.5.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 679 |
+
"visual.blocks.5.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 680 |
+
"visual.blocks.5.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 681 |
+
"visual.blocks.6.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 682 |
+
"visual.blocks.6.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 683 |
+
"visual.blocks.6.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 684 |
+
"visual.blocks.6.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 685 |
+
"visual.blocks.6.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 686 |
+
"visual.blocks.6.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 687 |
+
"visual.blocks.6.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 688 |
+
"visual.blocks.6.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 689 |
+
"visual.blocks.6.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 690 |
+
"visual.blocks.6.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 691 |
+
"visual.blocks.6.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 692 |
+
"visual.blocks.6.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 693 |
+
"visual.blocks.7.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 694 |
+
"visual.blocks.7.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 695 |
+
"visual.blocks.7.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 696 |
+
"visual.blocks.7.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 697 |
+
"visual.blocks.7.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 698 |
+
"visual.blocks.7.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 699 |
+
"visual.blocks.7.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 700 |
+
"visual.blocks.7.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 701 |
+
"visual.blocks.7.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 702 |
+
"visual.blocks.7.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 703 |
+
"visual.blocks.7.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 704 |
+
"visual.blocks.7.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 705 |
+
"visual.blocks.8.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 706 |
+
"visual.blocks.8.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 707 |
+
"visual.blocks.8.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 708 |
+
"visual.blocks.8.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 709 |
+
"visual.blocks.8.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 710 |
+
"visual.blocks.8.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 711 |
+
"visual.blocks.8.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 712 |
+
"visual.blocks.8.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 713 |
+
"visual.blocks.8.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 714 |
+
"visual.blocks.8.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 715 |
+
"visual.blocks.8.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 716 |
+
"visual.blocks.8.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 717 |
+
"visual.blocks.9.attn.proj.bias": "pytorch_model-00001-of-00004.bin",
|
| 718 |
+
"visual.blocks.9.attn.proj.weight": "pytorch_model-00001-of-00004.bin",
|
| 719 |
+
"visual.blocks.9.attn.qkv.bias": "pytorch_model-00001-of-00004.bin",
|
| 720 |
+
"visual.blocks.9.attn.qkv.weight": "pytorch_model-00001-of-00004.bin",
|
| 721 |
+
"visual.blocks.9.mlp.fc1.bias": "pytorch_model-00001-of-00004.bin",
|
| 722 |
+
"visual.blocks.9.mlp.fc1.weight": "pytorch_model-00001-of-00004.bin",
|
| 723 |
+
"visual.blocks.9.mlp.fc2.bias": "pytorch_model-00001-of-00004.bin",
|
| 724 |
+
"visual.blocks.9.mlp.fc2.weight": "pytorch_model-00001-of-00004.bin",
|
| 725 |
+
"visual.blocks.9.norm1.bias": "pytorch_model-00001-of-00004.bin",
|
| 726 |
+
"visual.blocks.9.norm1.weight": "pytorch_model-00001-of-00004.bin",
|
| 727 |
+
"visual.blocks.9.norm2.bias": "pytorch_model-00001-of-00004.bin",
|
| 728 |
+
"visual.blocks.9.norm2.weight": "pytorch_model-00001-of-00004.bin",
|
| 729 |
+
"visual.merger.ln_q.bias": "pytorch_model-00001-of-00004.bin",
|
| 730 |
+
"visual.merger.ln_q.weight": "pytorch_model-00001-of-00004.bin",
|
| 731 |
+
"visual.merger.mlp.0.bias": "pytorch_model-00001-of-00004.bin",
|
| 732 |
+
"visual.merger.mlp.0.weight": "pytorch_model-00001-of-00004.bin",
|
| 733 |
+
"visual.merger.mlp.2.bias": "pytorch_model-00001-of-00004.bin",
|
| 734 |
+
"visual.merger.mlp.2.weight": "pytorch_model-00001-of-00004.bin",
|
| 735 |
+
"visual.patch_embed.proj.weight": "pytorch_model-00001-of-00004.bin"
|
| 736 |
+
}
|
| 737 |
+
}
|
sam2_reducer_and_vlm_layer.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a3b4e62dbb45ad2e91c3fb3e3186b1cc2af258c60bec4656db08876dc88a097
|
| 3 |
+
size 8145862
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>",
|
| 5 |
+
"<|object_ref_start|>",
|
| 6 |
+
"<|object_ref_end|>",
|
| 7 |
+
"<|box_start|>",
|
| 8 |
+
"<|box_end|>",
|
| 9 |
+
"<|quad_start|>",
|
| 10 |
+
"<|quad_end|>",
|
| 11 |
+
"<|vision_start|>",
|
| 12 |
+
"<|vision_end|>",
|
| 13 |
+
"<|vision_pad|>",
|
| 14 |
+
"<|image_pad|>",
|
| 15 |
+
"<|video_pad|>"
|
| 16 |
+
],
|
| 17 |
+
"eos_token": {
|
| 18 |
+
"content": "<|im_end|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
},
|
| 24 |
+
"pad_token": {
|
| 25 |
+
"content": "<|endoftext|>",
|
| 26 |
+
"lstrip": false,
|
| 27 |
+
"normalized": false,
|
| 28 |
+
"rstrip": false,
|
| 29 |
+
"single_word": false
|
| 30 |
+
}
|
| 31 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:091aa7594dc2fcfbfa06b9e3c22a5f0562ac14f30375c13af7309407a0e67b8a
|
| 3 |
+
size 11420371
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,144 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"151643": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"151644": {
|
| 13 |
+
"content": "<|im_start|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"151645": {
|
| 21 |
+
"content": "<|im_end|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"151646": {
|
| 29 |
+
"content": "<|object_ref_start|>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": false,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"151647": {
|
| 37 |
+
"content": "<|object_ref_end|>",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
},
|
| 44 |
+
"151648": {
|
| 45 |
+
"content": "<|box_start|>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": true
|
| 51 |
+
},
|
| 52 |
+
"151649": {
|
| 53 |
+
"content": "<|box_end|>",
|
| 54 |
+
"lstrip": false,
|
| 55 |
+
"normalized": false,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": true
|
| 59 |
+
},
|
| 60 |
+
"151650": {
|
| 61 |
+
"content": "<|quad_start|>",
|
| 62 |
+
"lstrip": false,
|
| 63 |
+
"normalized": false,
|
| 64 |
+
"rstrip": false,
|
| 65 |
+
"single_word": false,
|
| 66 |
+
"special": true
|
| 67 |
+
},
|
| 68 |
+
"151651": {
|
| 69 |
+
"content": "<|quad_end|>",
|
| 70 |
+
"lstrip": false,
|
| 71 |
+
"normalized": false,
|
| 72 |
+
"rstrip": false,
|
| 73 |
+
"single_word": false,
|
| 74 |
+
"special": true
|
| 75 |
+
},
|
| 76 |
+
"151652": {
|
| 77 |
+
"content": "<|vision_start|>",
|
| 78 |
+
"lstrip": false,
|
| 79 |
+
"normalized": false,
|
| 80 |
+
"rstrip": false,
|
| 81 |
+
"single_word": false,
|
| 82 |
+
"special": true
|
| 83 |
+
},
|
| 84 |
+
"151653": {
|
| 85 |
+
"content": "<|vision_end|>",
|
| 86 |
+
"lstrip": false,
|
| 87 |
+
"normalized": false,
|
| 88 |
+
"rstrip": false,
|
| 89 |
+
"single_word": false,
|
| 90 |
+
"special": true
|
| 91 |
+
},
|
| 92 |
+
"151654": {
|
| 93 |
+
"content": "<|vision_pad|>",
|
| 94 |
+
"lstrip": false,
|
| 95 |
+
"normalized": false,
|
| 96 |
+
"rstrip": false,
|
| 97 |
+
"single_word": false,
|
| 98 |
+
"special": true
|
| 99 |
+
},
|
| 100 |
+
"151655": {
|
| 101 |
+
"content": "<|image_pad|>",
|
| 102 |
+
"lstrip": false,
|
| 103 |
+
"normalized": false,
|
| 104 |
+
"rstrip": false,
|
| 105 |
+
"single_word": false,
|
| 106 |
+
"special": true
|
| 107 |
+
},
|
| 108 |
+
"151656": {
|
| 109 |
+
"content": "<|video_pad|>",
|
| 110 |
+
"lstrip": false,
|
| 111 |
+
"normalized": false,
|
| 112 |
+
"rstrip": false,
|
| 113 |
+
"single_word": false,
|
| 114 |
+
"special": true
|
| 115 |
+
}
|
| 116 |
+
},
|
| 117 |
+
"additional_special_tokens": [
|
| 118 |
+
"<|im_start|>",
|
| 119 |
+
"<|im_end|>",
|
| 120 |
+
"<|object_ref_start|>",
|
| 121 |
+
"<|object_ref_end|>",
|
| 122 |
+
"<|box_start|>",
|
| 123 |
+
"<|box_end|>",
|
| 124 |
+
"<|quad_start|>",
|
| 125 |
+
"<|quad_end|>",
|
| 126 |
+
"<|vision_start|>",
|
| 127 |
+
"<|vision_end|>",
|
| 128 |
+
"<|vision_pad|>",
|
| 129 |
+
"<|image_pad|>",
|
| 130 |
+
"<|video_pad|>"
|
| 131 |
+
],
|
| 132 |
+
"bos_token": null,
|
| 133 |
+
"clean_up_tokenization_spaces": false,
|
| 134 |
+
"eos_token": "<|im_end|>",
|
| 135 |
+
"errors": "replace",
|
| 136 |
+
"extra_special_tokens": {},
|
| 137 |
+
"model_max_length": 32768,
|
| 138 |
+
"pad_token": "<|endoftext|>",
|
| 139 |
+
"padding_side": "left",
|
| 140 |
+
"processor_class": "Qwen2VLProcessor",
|
| 141 |
+
"split_special_tokens": false,
|
| 142 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 143 |
+
"unk_token": null
|
| 144 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|