|
--- |
|
license: apache-2.0 |
|
tags: |
|
- moe |
|
- merge |
|
language: |
|
- en |
|
- tr |
|
--- |
|
|
|
# Megatron_v3_2x7B |
|
|
|
Megatron_v3_2x7B is a bilingual Mixure of Experts (MoE) which can comprehend and speak English/Turkish. |
|
Megatron, MoE mimarisine sahip Türkçe ve İngilizce talimatları anlayan ve cevap veren bir modeldir. |
|
|
|
|
|
## 💻 Usage/Kullanımı |
|
|
|
```python |
|
!pip install -qU transformers bitsandbytes accelerate |
|
|
|
from transformers import AutoTokenizer |
|
import transformers |
|
import torch |
|
|
|
model = "Eurdem/Megatron_v3_2x7B" |
|
|
|
tokenizer = AutoTokenizer.from_pretrained(model) |
|
pipeline = transformers.pipeline( |
|
"text-generation", |
|
model=model, |
|
model_kwargs={"torch_dtype": torch.float16, "load_in_4bit": True}, |
|
) |
|
|
|
messages = [{"role": "user", "content": "Tell me about AI"}] |
|
prompt = pipeline.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) |
|
outputs = pipeline(prompt, max_new_tokens=1024, do_sample=True, temperature=0.7, top_k=500, top_p=0.95) |
|
print(outputs[0]["generated_text"]) |
|
``` |
