---
base_model:
- TeeZee/Kyllene-34B-v1.1
- Doctor-Shotgun/Nous-Capybara-limarpv3-34B
library_name: transformers
tags:
- mergekit
- merge
license: apache-2.0
language:
- en
---
# Kyllima 34B v1


## Model Details

This is a simple 50/50 merge of two of my favorite Yi 34B-based models for roleplay and creative writing, created using [mergekit](https://github.com/cg123/mergekit) on [Arcee.ai](https://app.arcee.ai/). 
200k context. Uncensored. 

Use with Alpaca-LimaRP or Vicuna prompt format.

Recommended settings: 0.8-1 temp, 1.1 rep pen, 1 top P, 0.05 min P, 40 top K. 

Add \</s> and \[INST] to stop strings. 

Adding instructions to write in complete sentences using correct spelling, grammar, punctuation, and spacing to the system prompt may help to eliminate extraneous spaces between punctuation and closing quotation marks.

The original upload had a broken tokenizer. If you downloaded before 10/2/24, please re-download.

Static GGUF available [here](https://huggingface.co/sirmyrrh/Kyllima-34B-v1-GGUF) or [here](https://huggingface.co/mradermacher/Kyllima-34B-v1-GGUF).
Imatrix GGUF available [here](https://huggingface.co/mradermacher/Kyllima-34B-v1-i1-GGUF). With thanks to [mradermacher](https://huggingface.co/mradermacher/).

### Merge Method

This model was merged using the [TIES](https://arxiv.org/abs/2306.01708) merge method using [TeeZee/Kyllene-34B-v1.1](https://huggingface.co/TeeZee/Kyllene-34B-v1.1) as a base.

### Models Merged

The following models were included in the merge:
* [TeeZee/Kyllene-34B-v1.1](https://huggingface.co/TeeZee/Kyllene-34B-v1.1)
* [Doctor-Shotgun/Nous-Capybara-limarpv3-34B](https://huggingface.co/Doctor-Shotgun/Nous-Capybara-limarpv3-34B)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
base_model: TeeZee/Kyllene-34B-v1.1
chat_template: auto
dtype: float16
merge_method: ties
models:
- model: TeeZee/Kyllene-34B-v1.1
  parameters:
    density: 0.5
    weight: 0.5
- model: Doctor-Shotgun/Nous-Capybara-limarpv3-34B
  parameters:
    density: 0.5
    weight: 0.5
parameters:
  embed_slerp: true
  int8_mask: true
  normalize: false
tokenizer_source: base

```