README.md

---
license: llama3
language:
- gsw
datasets:
- cis-lmu/Glot500
- cis-lmu/GlotCC-V1
pipeline_tag: text-generation
base_model: NousResearch/Hermes-2-Pro-Llama-3-8B
model_type: LlamaForCausalLM
tags:
- Llama-3
- instruct
- finetune
- qlora
- chatml
- synthetic data
- axolotl
---

# Alpesteibock-Llama-3-8B-Alpha

**Alpesteibock-Llama-3-8B-Alpha** is an experimental QLoRA fine-tune of [NousResearch/Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Pro-Llama-3-8B) on a dataset of more than 28 million tokens of Swiss German text from multiple sources.

## License

This model is released under the [Llama 3 Community License](https://llama.meta.com/llama3/license/).

## Usage

The model uses ChatML as an instruction template and was trained using "You are Alpesteibock, a helpful assistant who speaks Swiss German." as a system message:
```
<|im_start|>system
You are Alpesteibock, a helpful assistant who speaks Swiss German.<|im_end|>
<|im_start|>user
Hoi. Wie heissisch du?<|im_end|>
<|im_start|>assistant
Ich bi de Alpesteibock und ich freu mi uf di.<|im_end|>
```

## Dataset

The dataset used for training consists of the following sources:

| Dataset | File Size | Description | Phase |
|---------|-----------|-------------|-------|
| [Glot500 Corpus](https://huggingface.co/datasets/cis-lmu/Glot500) (gsw_Latn, Leipzig_web) | 21.7 MB | Text, usually sentences, crawled from the web | 1 |
| [Alemannic Wikipedia](https://dumps.wikimedia.org/alswiki/) (Subset) | 50.5 MB | Articles in the Alemannic Wikipedia with most of those written in Alsatian filtered out | 2 |
| [Schweizerdeutscher Mundartkorpus](https://chmk.ch/) (Copyright Free Subset) | 28.4 MB | Copyright free books written in Swiss German | 2 |
| [GlotCC-V1.0](https://huggingface.co/datasets/cis-lmu/GlotCC-V1) (gsw-Latn) | 7.5 MB | Document-level general domain monolingual dataset derived from CommonCrawl | 2 |
| Synthetic Instruction Data | 1.7 MB | Different datasets of synthetically generated Swiss German text | 2 |

## Training Details

Hardware: 1x RTX 4090  
Duration: 40 hours in total (2 hours for first phase and 38 hours for second phase)  

### Hyperparameters

Adapter: QLoRA  
Precision: 4-bit  
Optimizer: adamw_bnb_8bit  
LoRA Rank: 256  
LoRA Alpha: 256  
Learning Rate: 1e-5  
Scheduler: Cosine  
Context Length: 4096  
Batch Size: 1  
Gradient Accumulation Steps: 1  
Sample Packing: On for first phase, Off for second phase  
Epochs: 2