---
base_model:
- Undi95/Meta-Llama-3-8B-Instruct-hf
library_name: transformers
tags:
- mergekit
- merge

---
# Llama-3-8B-Ultra-Instruct

This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).

## Merge Details
### Merge Method

This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [Undi95/Meta-Llama-3-8B-Instruct-hf](https://huggingface.co./Undi95/Meta-Llama-3-8B-Instruct-hf) as a base.

### Models Merged

The following models were included in the merge:
* llama-3-8B-ultra-instruct/InstructPart
* llama-3-8B-ultra-instruct/RPPart

### Configuration

The following YAML configuration was used to produce this model:

```yaml
base_model: Undi95/Meta-Llama-3-8B-Instruct-hf
dtype: bfloat16
merge_method: dare_ties
slices:
- sources:
  - layer_range: [0, 32]
    model: llama-3-8B-ultra-instruct/RPPart
    parameters:
      weight: 0.39
  - layer_range: [0, 32]
    model: llama-3-8B-ultra-instruct/InstructPart
    parameters:
      weight: 0.26
  - layer_range: [0, 32]
    model: Undi95/Meta-Llama-3-8B-Instruct-hf
```