File size: 1,812 Bytes
931ea96 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 | # Combined Task Vector Model
This model was created by combining task vectors from multiple fine-tuned models.
## Task Vector Computation
```python
t_1 = TaskVector("Qwen/Qwen2.5-Coder-7B-Instruct", "/lustre10/scratch/tkwang/SecSteer/axolotl-outputs/lora-merged/plus-step-90")
t_2 = TaskVector("Qwen/Qwen2.5-Coder-7B-Instruct", "/lustre10/scratch/tkwang/SecSteer/axolotl-outputs/lora-merged/minus-step-90")
t_2 = TaskVector("Qwen/Qwen2.5-Coder-7B-Instruct", "None")
t_combined = 1.0 * t_1 + -1.0 * t_2
new_model = t_combined.apply_to("Qwen/Qwen2.5-Coder-7B-Instruct", scaling_coef=0.3)
```
Models Used
- Base Model: https://huggingface.co/Qwen/Qwen2.5-Coder-7B-Instruct
- Fine-tuned Model 1: https://huggingface.co//lustre10/scratch/tkwang/SecSteer/axolotl-outputs/lora-merged/plus-step-90
- Fine-tuned Model 2: https://huggingface.co//lustre10/scratch/tkwang/SecSteer/axolotl-outputs/lora-merged/minus-step-90
- Fine-tuned Model 3: https://huggingface.co/None
Technical Details
- Creation Script Git Hash: fb62f919e9796b294f1ffb6297b05d11fa945ac0
- Task Vector Method: Additive combination
- Args: {
"pretrained_model": "Qwen/Qwen2.5-Coder-7B-Instruct",
"finetuned_model1": "/lustre10/scratch/tkwang/SecSteer/axolotl-outputs/lora-merged/plus-step-90",
"finetuned_model2": "/lustre10/scratch/tkwang/SecSteer/axolotl-outputs/lora-merged/minus-step-90",
"finetuned_model3": null,
"apply_to_diff_model_architecure": null,
"output_model_name": "felixwangg/Qwen2.5-Coder-7B-Instruct-cpp-sec-step90-lam0.3",
"output_dir": "/lustre10/scratch/tkwang/SecSteer/axolotl-outputs/weight-arithmetic/step-90/lambda-0.3",
"scaling_coef": 0.3,
"apply_line_scaling_t1": false,
"apply_line_scaling_t2": false,
"apply_line_scaling_t3": false,
"scale_t1": 1.0,
"scale_t2": -1.0,
"scale_t3": null
}
|