---
base_model:
- MrRobotoAI/Thor-v1.4-8b-DARK-FICTION
- MrRobotoAI/8b-unaligned-BASE-v2d
- MrRobotoAI/8b-unaligned-BASE-v3e
- MrRobotoAI/HEL-v0.8-8b-LONG-DARK
- aifeifei798/llama3-8B-DarkIdol-2.3-Uncensored-32K
library_name: transformers
tags:
- mergekit
- merge

---
# merge

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 [TIES](https://arxiv.org/abs/2306.01708) merge method using [MrRobotoAI/Thor-v1.4-8b-DARK-FICTION](https://huggingface.co/MrRobotoAI/Thor-v1.4-8b-DARK-FICTION) as a base.

### Models Merged

The following models were included in the merge:
* [MrRobotoAI/8b-unaligned-BASE-v2d](https://huggingface.co/MrRobotoAI/8b-unaligned-BASE-v2d)
* [MrRobotoAI/8b-unaligned-BASE-v3e](https://huggingface.co/MrRobotoAI/8b-unaligned-BASE-v3e)
* [MrRobotoAI/HEL-v0.8-8b-LONG-DARK](https://huggingface.co/MrRobotoAI/HEL-v0.8-8b-LONG-DARK)
* [aifeifei798/llama3-8B-DarkIdol-2.3-Uncensored-32K](https://huggingface.co/aifeifei798/llama3-8B-DarkIdol-2.3-Uncensored-32K)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
merge_method: ties
models:
  - model: aifeifei798/llama3-8B-DarkIdol-2.3-Uncensored-32K
    parameters:
      weight:
        - filter: v_proj
          value: [0.5, 0.5, 0.49, 0.01, 0.01, 0.01, 0.01, 0.01, 0.49, 0.5, 0.5]
        - filter: o_proj
          value: [0.5, 0.5, 0.49, 0.01, 0.01, 0.01, 0.01, 0.01, 0.49, 0.5, 0.5]
        - filter: up_proj
          value: [0.5, 0.5, 0.49, 0.01, 0.01, 0.01, 0.01, 0.01, 0.49, 0.5, 0.5]
        - filter: gate_proj
          value: [0.5, 0.5, 0.49, 0.01, 0.01, 0.01, 0.01, 0.01, 0.49, 0.5, 0.5]
        - filter: down_proj
          value: [0.5, 0.5, 0.49, 0.01, 0.01, 0.01, 0.01, 0.01, 0.49, 0.5, 0.5]
        - value: 4
  - model: MrRobotoAI/Thor-v1.4-8b-DARK-FICTION
    parameters:
      weight:
        - filter: v_proj
          value: [0.5, 0.5, 0.49, 0.01, 0.01, 0.01, 0.01, 0.01, 0.49, 0.5, 0.5]
        - filter: o_proj
          value: [0.5, 0.5, 0.49, 0.01, 0.01, 0.01, 0.01, 0.01, 0.49, 0.5, 0.5]
        - filter: up_proj
          value: [0.5, 0.5, 0.49, 0.01, 0.01, 0.01, 0.01, 0.01, 0.49, 0.5, 0.5]
        - filter: gate_proj
          value: [0.5, 0.5, 0.49, 0.01, 0.01, 0.01, 0.01, 0.01, 0.49, 0.5, 0.5]
        - filter: down_proj
          value: [0.5, 0.5, 0.49, 0.01, 0.01, 0.01, 0.01, 0.01, 0.49, 0.5, 0.5]
        - value: 3
  - model: MrRobotoAI/HEL-v0.8-8b-LONG-DARK
    parameters:
      weight:
        - filter: v_proj
          value: [0, 0, 0.01, 0.48, 0.48, 0.48, 0.48, 0.48, 0.01, 0, 0]
        - filter: o_proj
          value: [0, 0, 0.01, 0.48, 0.48, 0.48, 0.48, 0.48, 0.01, 0, 0]
        - filter: up_proj
          value: [0, 0, 0.01, 0.48, 0.48, 0.48, 0.48, 0.48, 0.01, 0, 0]
        - filter: gate_proj
          value: [0, 0, 0.01, 0.48, 0.48, 0.48, 0.48, 0.48, 0.01, 0, 0]
        - filter: down_proj
          value: [0, 0, 0.01, 0.48, 0.48, 0.48, 0.48, 0.48, 0.01, 0, 0]
        - value: 2
  - model: MrRobotoAI/8b-unaligned-BASE-v3e
    parameters:
      weight:
        - filter: v_proj
          value: [0, 0, 0.01, 0.48, 0.48, 0.48, 0.48, 0.48, 0.01, 0, 0]
        - filter: o_proj
          value: [0, 0, 0.01, 0.48, 0.48, 0.48, 0.48, 0.48, 0.01, 0, 0]
        - filter: up_proj
          value: [0, 0, 0.01, 0.48, 0.48, 0.48, 0.48, 0.48, 0.01, 0, 0]
        - filter: gate_proj
          value: [0, 0, 0.01, 0.48, 0.48, 0.48, 0.48, 0.48, 0.01, 0, 0]
        - filter: down_proj
          value: [0, 0, 0.01, 0.48, 0.48, 0.48, 0.48, 0.48, 0.01, 0, 0]
        - value: 1
  - model: MrRobotoAI/8b-unaligned-BASE-v2d
    parameters:
      weight:
        - filter: v_proj
          value: [0, 0, 0, 0.02, 0.02, 0.02, 0.02, 0.02, 0, 0, 0]
        - filter: o_proj
          value: [0, 0, 0, 0.02, 0.02, 0.02, 0.02, 0.02, 0, 0, 0]
        - filter: up_proj
          value: [0, 0, 0, 0.02, 0.02, 0.02, 0.02, 0.02, 0, 0, 0]
        - filter: gate_proj
          value: [0, 0, 0, 0.02, 0.02, 0.02, 0.02, 0.02, 0, 0, 0]
        - filter: down_proj
          value: [0, 0, 0, 0.02, 0.02, 0.02, 0.02, 0.02, 0, 0, 0]
        - value: 0
base_model: MrRobotoAI/Thor-v1.4-8b-DARK-FICTION
tokenizer_source: base
dtype: bfloat16
```