|
--- |
|
license: apache-2.0 |
|
datasets: |
|
- YeungNLP/firefly-train-1.1M |
|
- LooksJuicy/ruozhiba |
|
language: |
|
- zh |
|
--- |
|
# Model Card for Llama 3 8B Instruct (Quantized to 4-bit) |
|
|
|
This model is a fine-tuned version of Llama 3 8B Instruct, quantized to 4-bit, using the Chinese datasets YeungNLP/firefly-train-1.1M and LooksJuicy/ruozhiba. |
|
|
|
## Model Details |
|
|
|
### Model Description |
|
|
|
- **Developed by:** Zane |
|
- **Model type:** Llama 3 8B Instruct (Quantized to 4-bit) |
|
- **Language(s) (NLP):** Chinese (zh) |
|
- **License:** Apache-2.0 |
|
|
|
## How to Get Started with the Model |
|
|
|
```python |
|
from transformers import AutoModelForCausalLM, AutoTokenizer |
|
|
|
model_name = "your-username/llama-3-8b-instruct-4bit-chinese" |
|
tokenizer = AutoTokenizer.from_pretrained(model_name) |
|
model = AutoModelForCausalLM.from_pretrained(model_name) |
|
|
|
input_text = "请输入您的中文文本" |
|
inputs = tokenizer(input_text, return_tensors="pt") |
|
outputs = model.generate(inputs.input_ids, max_length=50) |
|
generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True) |
|
print(generated_text) |
|
``` |
