54 lines
1.4 KiB
Markdown
54 lines
1.4 KiB
Markdown
---
|
|
license: cc-by-nc-4.0
|
|
---
|
|
|
|
Merge:
|
|
```shell
|
|
layer_slices:
|
|
- model: Gryphe/MythoMax-L2-13b
|
|
start: 0
|
|
end: 16
|
|
- model: Undi95/MM-ReMM-L2-20B-Part1
|
|
start: 8
|
|
end: 20
|
|
- model: Gryphe/MythoMax-L2-13b
|
|
start: 17
|
|
end: 32
|
|
- model: Undi95/MM-ReMM-L2-20B-Part1
|
|
start: 21
|
|
end: 40
|
|
```
|
|
|
|
<!-- description start -->
|
|
## Models used
|
|
|
|
- Gryphe/MythoMax-L2-13b
|
|
- Undi95/ReMM-v2.1-L2-13B
|
|
<!-- description end -->
|
|
|
|
Part1 = ReMM v2.1 merged /w MythoMax low weight to keep consistency. I call this "dilution" and result show consistency and coherency without repeat/loop beside the small amount of duplicated datas.
|
|
|
|
## Prompt template: Alpaca
|
|
|
|
```
|
|
Below is an instruction that describes a task. Write a response that completes the request.
|
|
|
|
### Instruction:
|
|
{prompt}
|
|
|
|
### Response:
|
|
```
|
|
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
|
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Undi95__MM-ReMM-L2-20B)
|
|
|
|
| Metric | Value |
|
|
|-----------------------|---------------------------|
|
|
| Avg. | 51.14 |
|
|
| ARC (25-shot) | 60.84 |
|
|
| HellaSwag (10-shot) | 85.18 |
|
|
| MMLU (5-shot) | 56.45 |
|
|
| TruthfulQA (0-shot) | 53.33 |
|
|
| Winogrande (5-shot) | 75.77 |
|
|
| GSM8K (5-shot) | 7.73 |
|
|
| DROP (3-shot) | 18.66 |
|