36 lines
810 B
Markdown
36 lines
810 B
Markdown
|
|
---
|
||
|
|
license: apache-2.0
|
||
|
|
tags:
|
||
|
|
- merge
|
||
|
|
- mergekit
|
||
|
|
- axiong/PMC_LLaMA_13B
|
||
|
|
- lmsys/vicuna-13b-v1.3
|
||
|
|
---
|
||
|
|
|
||
|
|
# arcee-ai/PMC_LLaMA_Vicuna_13B_Slerp
|
||
|
|
|
||
|
|
arcee-ai/PMC_LLaMA_Vicuna_13B_Slerp is a merge of the following models using [mergekit](https://github.com/cg123/mergekit):
|
||
|
|
* [axiong/PMC_LLaMA_13B](https://huggingface.co/axiong/PMC_LLaMA_13B)
|
||
|
|
* [lmsys/vicuna-13b-v1.3](https://huggingface.co/lmsys/vicuna-13b-v1.3)
|
||
|
|
|
||
|
|
## 🧩 Configuration
|
||
|
|
|
||
|
|
```yaml
|
||
|
|
slices:
|
||
|
|
- sources:
|
||
|
|
- model: axiong/PMC_LLaMA_13B
|
||
|
|
layer_range: [0, 40]
|
||
|
|
- model: lmsys/vicuna-13b-v1.3
|
||
|
|
layer_range: [0, 40]
|
||
|
|
merge_method: slerp
|
||
|
|
base_model: lmsys/vicuna-13b-v1.3
|
||
|
|
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
|
||
|
|
|
||
|
|
```
|