|
--- |
|
license: apache-2.0 |
|
datasets: |
|
- yahma/alpaca-cleaned |
|
language: |
|
- de |
|
tags: |
|
- llama |
|
- alpaca |
|
- ggml |
|
- german |
|
- deutsch |
|
- zicklein |
|
--- |
|
|
|
# --- |
|
|
|
--- |
|
|
|
|
|
|
|
# Zicklein: A german finetuned instructions following LLaMA |
|
|
|
|
|
## This is a ggml conversion of [Zicklein](https://github.com/avocardio/zicklein) 7B. |
|
|
|
## Zicklein itself is a LLaMA finetuned model with a cleaned and german translated [Alpaca](https://github.com/tatsu-lab/stanford_alpaca) [dataset](https://github.com/LEL-A/GerAlpacaDataCleaned). |
|
|
|
Currently I have only converted it into **new k-quant method Q5_K_M**. I will gladly make more versions on request. |
|
|
|
Other possible quantizations include: q2_K, q3_K_S, q3_K_M, q3_K_L, q4_K_S, q4_K_M, q5_K_S, q5_K_M, q6_K |
|
|
|
A f-16 version could be found here: [nikuya3/alpaca-lora-7b-german-base-51k-ggml](https://huggingface.co/nikuya3/alpaca-lora-7b-german-base-51k-ggml) |
|
|
|
Compatible with **llama.cpp**, but also with: |
|
|
|
- **text-generation-webui** |
|
- **KoboldCpp** |
|
- **ParisNeo/GPT4All-UI** |
|
- **llama-cpp-python** |
|
- **ctransformers** |
|
|
|
--- |
|
|
|
## Prompt format |
|
|
|
Since this model is based on alpaca dataset, the right prompt formatting should look like this: |
|
|
|
``` |
|
Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request. |
|
|
|
### Instruction: |
|
{instruction} |
|
|
|
### Input: |
|
{input} |
|
|
|
### Response: |
|
``` |
|
|
|
|
|
Or **without** addiotional **input**: |
|
|
|
|
|
``` |
|
|
|
Below is an instruction that describes a task. Write a response that appropriately completes the request. |
|
|
|
### Instruction: |
|
{instruction} |
|
|
|
### Response: |
|
``` |
|
|
|
--- |
|
|
|
### That's it! |
|
|
|
|
|
If you have any further questions, feel free to contact me or start a discussion |
