38 lines
962 B
Markdown
38 lines
962 B
Markdown
---
|
|
library_name: transformers
|
|
license: apache-2.0
|
|
---
|
|
|
|
## T3Q-MSlerp-7Bx2
|
|
|
|
T3Q-MSlerp-7Bx2 is a merge of the following models using [mergekit](https://github.com/cg123/mergekit):
|
|
* [zhengr/MixTAO-7Bx2-MoE-v8.1](https://huggingface.co/zhengr/MixTAO-7Bx2-MoE-v8.1)
|
|
* [yunconglong/Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B](https://huggingface.co/yunconglong/Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B)
|
|
|
|
|
|
**Model Developers** Chihoon Lee(chlee10), T3Q
|
|
|
|
|
|
```yaml
|
|
|
|
slices:
|
|
- sources:
|
|
- model: zhengr/MixTAO-7Bx2-MoE-v8.1
|
|
layer_range: [0, 32]
|
|
- model: yunconglong/Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B
|
|
layer_range: [0, 32]
|
|
|
|
merge_method: slerp
|
|
base_model: zhengr/MixTAO-7Bx2-MoE-v8.1
|
|
|
|
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 # fallback for rest of tensors
|
|
|
|
dtype: float16
|
|
|
|
``` |