trlx
RL4LMs
trlx | RL4LMs | |
---|---|---|
6 | 5 | |
4,332 | 2,094 | |
1.1% | 2.2% | |
7.9 | 0.0 | |
4 months ago | 2 months ago | |
Python | Python | |
MIT License | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
trlx
-
Recapping the AI, Machine Learning and Data Science Meetup — May 2, 2024
Transformer Reinforcement Learning X on GitHub
-
Why did Stability not copy Midjourney's RLHF process? And what's the future of Stable Diffusion?
We drove and released the top RLHF framework TRLX for example from our Carper AI lab used by some of the biggest companies in the world: https://github.com/CarperAI/trlx
-
[R] Stanford-Alpaca 7B model (an instruction tuned version of LLaMA) performs as well as text-davinci-003
If you checkout the trlx repo they have some examples and they have an example of how they trained sft and ppo on the hh dataset. So it’s basically that but with llama. https://github.com/CarperAI/trlx/blob/main/examples/hh/sft_hh.py
-
Sam Altman: OpenAI’s GPT-4 will launch only when they can do it safely & responsibly. “In general we are going to release technology much more slowly than people would like. We're going to sit on it for much longer…”. Also confirmed video model in the works.
We’ll release our first trained model with Stability AI soon. If you want to start tinkering with RLHF now, we’re also helping develop TRLX: https://github.com/CarperAI/trlx — the open source library for reinforcement learning with transformers.
-
[P] RLHF Learning to Summarize: Implementation by CarperAI with trlX
trlX library here: https://github.com/CarperAI/trlx
- Will we ever see an open source alternative to ChatGPT?
RL4LMs
-
How To Setup a Model With Guardrails?
I think of guardrails as another dimension of human preferences: whether you are training a model to answer questions more gooder or avoid saying horrifying stuff, you are teaching the model a preference. So I thinks it's a straightforward RLHF problem but from a different perspective.
-
OpenDILab Awesome Paper Collection: RL with Human Feedback (2)
Found relevant code at https://github.com/allenai/rl4lms + all code implementations here
-
Best option for creating a custom GPT AI
If you skip down ther s several open source options. https://github.com/allenai/RL4LMs is an example.
- Will we ever see an open source alternative to ChatGPT?
- An Open-Source Version of ChatGPT is Coming [News]
What are some alternatives?
alpaca-lora - Instruct-tune LLaMA on consumer hardware
Dromedary - Dromedary: towards helpful, ethical and reliable LLMs.
PaLM-rlhf-pytorch - Implementation of RLHF (Reinforcement Learning with Human Feedback) on top of the PaLM architecture. Basically ChatGPT but with PaLM
trl - Train transformer language models with reinforcement learning.
valhalla-nmt - Code repository for CVPR 2022 paper "VALHALLA: Visual Hallucination for Machine Translation"
text-generation-webui - A Gradio web UI for Large Language Models. Supports transformers, GPTQ, AWQ, EXL2, llama.cpp (GGUF), Llama models.
Open-Assistant - OpenAssistant is a chat-based assistant that understands tasks, can interact with third-party systems, and retrieve information dynamically to do so.
summarize-from-feedback - Code for "Learning to summarize from human feedback"
awesome-transformer-nlp - A curated list of NLP resources focused on Transformer networks, attention mechanism, GPT, BERT, ChatGPT, LLMs, and transfer learning.