---
base_model:
- unsloth/Meta-Llama-3.1-8B
- unsloth/Meta-Llama-3.1-8B-Instruct
- unsloth/Meta-Llama-3.1-8B
- KayaTechAI/llama-3.1-8b-financial-expert-pre-trained
library_name: transformers
tags:
- mergekit
- merge

---
# merged_financial_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 [Task Arithmetic](https://arxiv.org/abs/2212.04089) merge method using [unsloth/Meta-Llama-3.1-8B](https://huggingface.co/unsloth/Meta-Llama-3.1-8B) as a base.

### Models Merged

The following models were included in the merge:
* [unsloth/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/unsloth/Meta-Llama-3.1-8B-Instruct)
* [unsloth/Meta-Llama-3.1-8B](https://huggingface.co/unsloth/Meta-Llama-3.1-8B) + [KayaTechAI/llama-3.1-8b-financial-expert-pre-trained](https://huggingface.co/KayaTechAI/llama-3.1-8b-financial-expert-pre-trained)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
merge_method: task_arithmetic
base_model: unsloth/Meta-Llama-3.1-8B
rat_chet: false
dtype: bfloat16

models:
  # 1. Base model combined with your financial CPT adapter
  - model: unsloth/Meta-Llama-3.1-8B+KayaTechAI/llama-3.1-8b-financial-expert-pre-trained
    parameters:
      weight: 1.0

  # 2. Reinject the isolated Instruction alignment delta
  - model: unsloth/Meta-Llama-3.1-8B-Instruct
    parameters:
      weight: 0.35 # Standard weight to restore instruction-following without wiping domain facts

tokenizer_source: unsloth/Meta-Llama-3.1-8B-Instruct

```