axolotl
chat.petals.dev
axolotl | chat.petals.dev | |
---|---|---|
29 | 8 | |
5,811 | 298 | |
9.3% | 2.3% | |
9.8 | 7.1 | |
5 days ago | 12 days ago | |
Python | Python | |
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.
axolotl
-
Ask HN: Most efficient way to fine-tune an LLM in 2024?
The approach I see used is axolotl with QLoRA using cloud GPUs which can be quite cheap.
https://github.com/OpenAccess-AI-Collective/axolotl
- FLaNK AI - 01 April 2024
-
LoRA from Scratch implementation for LLM finetuning
https://github.com/OpenAccess-AI-Collective/axolotl
- Optimized Triton Kernels for full fine tunes
- Axolotl
-
Let’s Collaborate to Build a High-Quality, Open-Source Dataset for LLMs!
One option is to look at what Axolotl uses. They have a list of different dataset formats that they support. They're mostly in JSON with specific field names, so you could start putting a dataset together with a text editor or a JSON editor.
- Axolotl: Streamline fine-tuning of AI models
-
Dataset Creation Tools?
You can save that overall set into a json file and load it up as training data in whatever you're using. I'm using axolotl for it at the moment. Though a GUI based option is probably best for the first couple of tries until you get a feel for the options.
-
Progress on Reproducing Phi-1/1.5
Looking forward to the results! If it turns out the dataset is reproducible, then it might be a good candidate for ReLora training on axolotl!
chat.petals.dev
-
Make no mistake—AI is owned by Big Tech
ETA: https://chat.petals.dev
-
Run LLMs in bittorrent style
Check it out at Petals.dev. Chatbot
-
Run LLMs at home, BitTorrent‑style
Hi, a dev here. means "end of sequence" for LLMs. If a model generates it, it forgets everything and continue with an unrelated random text. So I don't think that malicious actors are involved here.
Apparently, the Colab code snippet is just too simplified and does not handle correctly. This is not the case with the full chatbot app at https://chat.petals.dev - you can use it instead or take a look at its code.
-
Falcon180B: authors open source a new 180B version!
edit: this community of people is amazing. like 10 minutes after I posted this or so.... it is now up on chat.petals.dev !!!!
- Talk to Falcon 180B-Chat running over Petals
-
ChatGPT Is Down Again
good opportunity to try the free and totally open source Big Science Petals chat: https://chat.petals.dev/ ... Try out Stable Beluga 2 70B
I am currently running my 3090 GPU on there to help out, you can check out https://health.petals.dev/
If you have a spare GPU, consider contributing: https://github.com/bigscience-workshop/petals . I am not associated with them.
-
Sweating Bullets Test
So far, not a single one of the models tested (between 7b-70b) could figure out the name of the main character (Nick Slaughter). I've tried all sorts of prompts and the connection between "Tropical Heat" and "Sweating Bullets" is usually known to the model (e.g. "What's the show "Tropical Heat" called in the US?"). But as soon as I ask about the main character, all the models I have tested so far hallucinate all sorts of names, though usually in the right direction (detectives).
- Petals: Run 100B+ models at home bit-torrent style
What are some alternatives?
signal-cli - signal-cli provides an unofficial commandline, JSON-RPC and dbus interface for the Signal messenger.
askai - Command Line Interface for OpenAi ChatGPT
gpt-llm-trainer
petals - 🌸 Run LLMs at home, BitTorrent-style. Fine-tuning and inference up to 10x faster than offloading
LoRA - Code for loralib, an implementation of "LoRA: Low-Rank Adaptation of Large Language Models"
ggml - Tensor library for machine learning
mlc-llm - Enable everyone to develop, optimize and deploy AI models natively on everyone's devices.
LMFlow - An Extensible Toolkit for Finetuning and Inference of Large Foundation Models. Large Models for All.
KoboldAI-Client
koboldcpp - A simple one-file way to run various GGML and GGUF models with KoboldAI's UI
ollama - Get up and running with Llama 3, Mistral, Gemma, and other large language models.