|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
base_model: yanolja/EEVE-Korean-2.8B-v1.0 |
|
|
|
--- |
|
<p align="left"> |
|
<img src="https://huggingface.co/Neversleep-3B-Instruct-v0.1/resolve/main/eeve_logo.webp" width="100%"/> |
|
<p> |
|
|
|
# "We must sleep, but AI Never Sleeps!" |
|
|
|
|
|
## Prompt Template |
|
``` |
|
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions. |
|
Human: {prompt} |
|
Assistant: |
|
``` |
|
## Simple-Usage |
|
```python |
|
from transformers import AutoTokenizer |
|
from transformers import AutoModelForCausalLM |
|
|
|
model = AutoModelForCausalLM.from_pretrained("yanolja/EEVE-Korean-Instruct-2.8B-v1.0", trust_remote_code=True) |
|
tokenizer = AutoTokenizer.from_pretrained("yanolja/EEVE-Korean-Instruct-2.8B-v1.0", trust_remote_code=True) |
|
|
|
prompt_template = "A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.\nHuman: {prompt}\nAssistant:\n" |
|
text = 'λ€μ΄μ΄νΈμ λ©λ΄λ₯Ό μΆμ²ν΄μ£ΌμΈμ.\n\n(A) μλ¬λ\n(B) μΉν¨\n(C) νΌμ\n(D) νμ€ν' |
|
model_inputs = tokenizer(prompt_template.format(prompt=text), return_tensors='pt') |
|
|
|
outputs = model.generate(**model_inputs, max_new_tokens=256) |
|
output_text = tokenizer.batch_decode(outputs, skip_special_tokens=True)[0] |
|
print(output_text) |
|
``` |
|
|
|
### Example Output |
|
``` |
|
A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions. |
|
Human: λ€μ΄μ΄νΈμ λ©λ΄λ₯Ό μΆμ²ν΄μ£ΌμΈμ. |
|
|
|
(A) μλ¬λ |
|
(B) μΉν¨ |
|
(C) νΌμ |
|
(D) νμ€ν |
|
Assistant: |
|
(A) μλ¬λλ₯Ό μΆμ²λ립λλ€. μλ¬λλ μ μΉΌλ‘리μ΄λ©΄μλ μμμκ° νλΆν΄ λ€μ΄μ΄νΈμμΌλ‘ μ ν©ν©λλ€. λ€μν μ±μμ λ¨λ°±μ§μ μΆκ°νμ¬ κ· ν μ‘ν μμ¬λ₯Ό λ§λμ€ μ μμ΅λλ€. |
|
``` |
|
|
|
## About the Model |
|
|
|
First of all, Overwhelming gratitude to 'yanolja/EEVE' Model & Team! |
|
This model is a fine-tuned version of [crimsonjoo/Neversleep-3B-v0.1](https://huggingface.co/crimsonjoo/Neversleep-3B-v0.1), which is a Korean vocabulary-extended version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2). Specifically, we utilized Direct Preference Optimization (DPO) through the use of [Axolotl](https://github.com/OpenAccess-AI-Collective/axolotl). |
|
|
|
For more details, please refer to our technical report: [Efficient and Effective Vocabulary Expansion Towards Multilingual Large Language Models](https://arxiv.org/abs/2402.14714). |
|
|
|
## Training Data |
|
- Korean-translated version of [Open-Orca/SlimOrca-Dedup](https://huggingface.co/datasets/Open-Orca/SlimOrca-Dedup) |
|
- Korean-translated version of [argilla/ultrafeedback-binarized-preferences-cleaned](https://huggingface.co/datasets/argilla/ultrafeedback-binarized-preferences-cleaned) |
|
- No other dataset was used |