48 lines
1.8 KiB
Markdown
48 lines
1.8 KiB
Markdown
|
---
|
||
|
|
license: apache-2.0
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
library_name: transformers
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
---
|
||
|
|
|
||
|
|
# Cyrax-7B
|
||
|
|
|
||
|
|
## 🏆 Evaluation
|
||
|
|
|
||
|
|
### Open LLM Leaderboard
|
||
|
|
|
||
|
|
| Model |Average|ARC|HellaSwag|MMLU|TruthfulQA|Winogrande|GSM8K
|
||
|
|
|------------------------------------------------------------|------:|------:|---------:|-------:|------:|------:|------:|
|
||
|
|
|[**Cyrax-7B**](https://huggingface.co/touqir/Cyrax-7B)| **75.98**| **72.95**| 88.19| 64.6| **77.01**| 83.9| **69.22** |
|
||
|
|
|[Qwen-72B](https://huggingface.co/Qwen/Qwen-72B)| 73.6| 65.19| 85.94| **77.37**| 60.19| 82.48| 70.43|
|
||
|
|
|[Mixtral-8x7B-Instruct-v0.1-DPO](https://huggingface.co/cloudyu/Mixtral-8x7B-Instruct-v0.1-DPO)| 73.44| 69.8| 87.83| 71.05| 69.18| 81.37| 61.41|
|
||
|
|
|[Mixtral-8x7B-Instruct-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1)| 72.7| 70.14 | 87.55| 71.4| 64.98| 81.06| 61.11 |
|
||
|
|
|[llama2_70b_mmlu](https://huggingface.co/itsliupeng/llama2_70b_mmlu)| 68.24| 65.61| 87.37| 71.89| 49.15| 82.4| 52.99 |
|
||
|
|
|[falcon-180B](https://huggingface.co/tiiuae/falcon-180B)| 67.85| 69.45| **88.86**| 70.5| 45.47| **86.9**| 45.94|
|
||
|
|
|
||
|
|
|
||
|
|
## 💻 Usage
|
||
|
|
|
||
|
|
```python
|
||
|
|
!pip install -qU transformers accelerate
|
||
|
|
|
||
|
|
from transformers import AutoTokenizer
|
||
|
|
import transformers
|
||
|
|
import torch
|
||
|
|
|
||
|
|
model = "touqir/Cyrax-7B"
|
||
|
|
messages = [{"role": "user", "content": "What is Huggingface?"}]
|
||
|
|
|
||
|
|
tokenizer = AutoTokenizer.from_pretrained(model)
|
||
|
|
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
|
||
|
|
pipeline = transformers.pipeline(
|
||
|
|
"text-generation",
|
||
|
|
model=model,
|
||
|
|
torch_dtype=torch.float16,
|
||
|
|
device_map="auto",
|
||
|
|
)
|
||
|
|
|
||
|
|
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
|
||
|
|
print(outputs[0]["generated_text"])
|
||
|
|
```
|