AlekseiPravdin's picture
Create README.md
8a8abb9 verified
|
raw
history blame
1.91 kB
---
license: apache-2.0
tags:
- merge
- mergekit
- lazymergekit
- Nitral-AI/Nyan-Stunna-7B
- Nitral-AI/Kunocchini-7b-128k-test
- gguf
- Q2_K
- Q3_K_L
- Q3_K_M
- Q3_K_S
- Q4_0
- Q4_1
- Q4_K_S
- Q4_k_m
- Q5_0
- Q5_1
- Q6_K
- Q5_K_S
- Q5_k_m
- Q8_0
- 128k
language:
- en
- ru
- th
---
# NSK-7B-128k-slerp
NSK-7B-128k-slerp is a merge of the following models using [mergekit](https://github.com/cg123/mergekit):
* [Nitral-AI/Nyan-Stunna-7B](https://huggingface.co/Nitral-AI/Nyan-Stunna-7B)
* [Nitral-AI/Kunocchini-7b-128k-test](https://huggingface.co/Nitral-AI/Kunocchini-7b-128k-test)
## 🧩 Configuration
```yaml
slices:
- sources:
- model: Nitral-AI/Nyan-Stunna-7B
layer_range: [0, 32]
- model: Nitral-AI/Kunocchini-7b-128k-test
layer_range: [0, 32]
merge_method: slerp
base_model: Nitral-AI/Kunocchini-7b-128k-test
parameters:
t:
- filter: self_attn
value: [0, 0.5, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.5, 0.7, 0.3, 0]
- value: 0.5
dtype: bfloat16
```
Eval embedding benchmark (with 70 specific quesions):
![inf.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/UbeMfW28pMHSRLsSbEsJB.jpeg)
![md28g.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/6UNV3CaKdofeAUr7C7x9k.jpeg)
![SK.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/uSnHhxDCqo9DP9oSb_l6j.jpeg)
![ks-inf.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/1ekTvK84ZlEsFFOYWOHE4.jpeg)
![command-r.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/5lVz28EK07RmrUe49y4jn.jpeg)
![NSK.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/aNdIdS5MnkwJ9YhprGznw.jpeg)
![NSMv2.jpg](https://cdn-uploads.huggingface.co/production/uploads/6404a7eaad54665351d89135/vk2GpfnJnYS5u1_wA1Nhr.jpeg)