|
--- |
|
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 |
|
- 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 |
