42 lines
887 B
Markdown
42 lines
887 B
Markdown
---
|
|
license: apache-2.0
|
|
tags:
|
|
- merge
|
|
- mergekit
|
|
- samir-fama/SamirGPT-v1
|
|
- mlabonne/NeuralHermes-2.5-Mistral-7B
|
|
- KoboldAI/Mistral-7B-Erebus-v3
|
|
---
|
|
|
|
# ErebusNeuralSamir-7B-dare-ties
|
|
|
|
ErebusNeuralSamir-7B-dare-ties is a merge of the following models using mergekit:
|
|
* samir-fama/SamirGPT-v1
|
|
* mlabonne/NeuralHermes-2.5-Mistral-7B
|
|
* KoboldAI/Mistral-7B-Erebus-v3
|
|
|
|
## 🧩 Configuration
|
|
|
|
\\\```yaml
|
|
models:
|
|
- model: mistralai/Mistral-7B-v0.1
|
|
# No parameters necessary for base model
|
|
- model: samir-fama/SamirGPT-v1
|
|
parameters:
|
|
density: 0.53
|
|
weight: 0.3
|
|
- model: mlabonne/NeuralHermes-2.5-Mistral-7B
|
|
parameters:
|
|
density: 0.53
|
|
weight: 0.3
|
|
- model: KoboldAI/Mistral-7B-Erebus-v3
|
|
parameters:
|
|
density: 0.53
|
|
weight: 0.4
|
|
merge_method: dare_ties
|
|
base_model: mistralai/Mistral-7B-v0.1
|
|
parameters:
|
|
int8_mask: true
|
|
dtype: bfloat16
|
|
|
|
\\\``` |