Views
No views yet
| Model | Arena-Hard |
|---|---|
| Claude-3.5-Sonnet (Proprietary) | 79.3% |
| GPT-4o (Proprietary) | 79.2% |
| Athene-70B (Open) | 77.8% |
| Gemini-Pro-1.5 (Proprietary) | 72.0% |
| Gemma-2-27B (Open) | 57.0% |
| Llama-3-70B (Open) | 46.6% |
1import transformers
2import torch
3
4model_id = "Nexusflow/Athene-70B"
5
6pipeline = transformers.pipeline(
7 "text-generation",
8 model=model_id,
9 model_kwargs={"torch_dtype": torch.bfloat16},
10 device_map="auto",
11)
12
13messages = [
14 {"role": "system", "content": "You are an Athene Noctura, you can only speak with owl sounds. Whoooo whooo."},
15 {"role": "user", "content": "Whooo are you?"},
16]
17
18terminators = [
19 pipeline.tokenizer.eos_token_id,
20 pipeline.tokenizer.convert_tokens_to_ids("<|end_of_text|>")
21]
22
23outputs = pipeline(
24 messages,
25 max_new_tokens=256,
26 eos_token_id=terminators,
27 do_sample=True,
28 temperature=0.6,
29 top_p=0.9,
30)
31print(outputs[0]["generated_text"][-1])@misc{Athene2024,
title = {Athene-70B: Redefining the Boundaries of Post-Training for Open Models},
url = {https://nexusflow.ai/blogs/athene},
author = {Frick, Evan and Jin, Peter and Li, Tianle and Ganesan, Karthik and Zhang, Jian and Jiao, Jiantao and Zhu, Banghua},
month = {July},
year = {2024}
}