Table of Contents
哪里玩原版 / LLaMA 2 Demo
- HuggingChat/Llama-2-70b-chat-hf
- Hugging Face Space 7B | 13B | 70B
- Perplexity 7B/13B
- Replicate 7B | 13B | 70B
- sdk.vercel.ai
- nat.dev
怎么玩 / How to Prompt
<s>[INST] <<SYS>>
{{ system_prompt }}
<</SYS>>
{{ user_message }} [/INST]
<s>[INST] <<SYS>>
{{ system_prompt }}
<</SYS>>
{{ user_msg_1 }} [/INST] {{ model_answer_1 }} </s>\
<s>[INST] {{ user_msg_2 }} [/INST] {{ model_answer_2 }} </s>\
<s>[INST] {{ user_msg_3 }} [/INST]
跑分 / Benchmark
细节 / Detail
- Llama 2: Open Foundation and Fine-Tuned Chat Models
- Llama 2: an incredible open LLM
- Llama 2: Full Breakdown
- https://twitter.com/DrJimFan/status/1681372700881854465
微调 / Fine-tuning
已经有哪些了？ / Where
- https://huggingface.co/spaces/LinkSoul/Chinese-Llama-2-7b 中文
- https://huggingface.co/NousResearch/Nous-Hermes-Llama2-13b Hermes
- https://huggingface.co/stabilityai/FreeWilly2 Orca-like
怎么调？/ How
- Fine-tune LLaMA 2 (7-70B) on Amazon SageMaker
- HuggingFace, Fine-tuning LLaMA 2 with PEFT
- Meta Examples and recipes for Llama model
- The EASIEST way to finetune LLAMA-v2 on local machine!
- 知乎，LLaMA2 RLHF 技术细节
- Ollama
其它 / Misc.
美西时间这周六早上九点，美东时间早上十二点，聊一下怎么把 LLaMA 2 搞到超过 GPT 3.5 的效果，有兴趣一起来讨论
Topic: LLaMA-2 搞到 match / 超过 GPT 3.5
Time: Jul 22, 2023 09:00 AM Pacific Time (US and Canada)
https://ed-ac-uk.zoom.us/j/87003419834
Meeting ID: 870 0341 9834
Passcode: abcd1234
批评
- https://twitter.com/MarcosGorgojo/status/1681365374606901255
- https://twitter.com/dotey/status/1681925505644101632
