ingridpan

11 ms·

1.
by ingridpan 3y ago
I found this tutorial helpful for getting started with fine-tuning https://www.youtube.com/watch?v=74NSDMvYZ9Y This guy used gradient.ai and he has a Google Collab to try it
2.
by ingridpan 3y ago
While you're experimenting, worth checking out https://gradient.ai/ -- they're basically the OpenAI API but with llama2
3.
by ingridpan 3y ago
Research: https://arxiv.org/abs/2005.11401 An implementation on Hugging Face: https://huggingface.co/docs/transformers/model_doc/rag
4.
by ingridpan 3y ago
In this post, we explore problems involved in LLM deployment, from GPU shortages to bottlenecks in model performance. These problems have inspired recent developments in distributed training frameworks commonly used to train LLMs, notably Z
5.
by ingridpan 3y ago
RAG is great for pulling some additional knowledge, but if you combine it with fine-tuning (i.e., the LLM 'understands' the domain-specific terminology better) it becomes a lot more effective
6.
by ingridpan 3y ago
https://gradient.ai/ API for inference and fine-tuning open-source LLMs
7.
by ingridpan 3y ago
Good question, esp as Gradient fine-tuning is so much cheaper than Open AI's
8.
by ingridpan 3y ago
not quite self-hosted but gradient.ai gives you access to llama2 via CLI
9.
by ingridpan 3y ago
https://gradient.ai/ is doing that with llama2
10.
Squeeze more out of your GPU for LLM inference–Accelerate and DeepSpeed tutorial(gradient.ai)
1 points by ingridpan 3y ago|0 comments
11.
Can We Prompt an LLM to Uncover Its Dreams of Electric Sheep?(gradient.ai)
2 points by ingridpan 3y ago|0 comments
12.
How to customize AI to solve your problems [video](youtube.com)
2 points by ingridpan 3y ago|0 comments
13.
Best practices for fine-tuning LLMs [video](youtube.com)
3 points by ingridpan 3y ago|0 comments