vllm
gpt-llm-trainer | vllm | |
---|---|---|
4 | 31 | |
3,814 | 19,344 | |
- | 12.6% | |
5.4 | 9.9 | |
about 2 months ago | 3 days ago | |
Jupyter Notebook | 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.
gpt-llm-trainer
- FLaNK Stack Weekly 06 Nov 2023
-
Show HN: Fine-tune your own Llama 2 to replace GPT-3.5/4
Very nice, thanks!
Check out what Matt Shumer put together as well: https://github.com/mshumer/gpt-llm-trainer.
I have used his trainer for auto distillation of GPT-4 into GPT3.5 fine tunes, but plan to do the same for Llama as well.
Cheers!
-
[D] Anyone tried gpt-llm-trainer?
Hey guys, so I stumbled upon this Linkedin post, this guy was showing a jupyter notebook on google colab and was explaining step by step how to train your own model to accomplish very specific tasks, and I believe the base model he was using Llama 2 7B Fine tuning version. This is the github link: https://github.com/mshumer/gpt-llm-trainer
- GPT-LLM-Trainer
vllm
-
AI leaderboards are no longer useful. It's time to switch to Pareto curves
I guess the root cause of my claim is that OpenAI won't tell us whether or not GPT-3.5 is an MoE model, and I assumed it wasn't. Since GPT-3.5 is clearly nondeterministic at temp=0, I believed the nondeterminism was due to FPU stuff, and this effect was amplified with GPT-4's MoE. But if GPT-3.5 is also MoE then that's just wrong.
What makes this especially tricky is that small models are truly 100% deterministic at temp=0 because the relative likelihoods are too coarse for FPU issues to be a factor. I had thought 3.5 was big enough that some of its token probabilities were too fine-grained for the FPU. But that's probably wrong.
On the other hand, it's not just GPT, there are currently floating-point difficulties in vllm which significantly affect the determinism of any model run on it: https://github.com/vllm-project/vllm/issues/966 Note that a suggested fix is upcasting to float32. So it's possible that GPT-3.5 is using an especially low-precision float and introducing nondeterminism by saving money on compute costs.
Sadly I do not have the money[1] to actually run a test to falsify any of this. It seems like this would be a good little research project.
[1] Or the time, or the motivation :) But this stuff is expensive.
-
Mistral AI Launches New 8x22B Moe Model
The easiest is to use vllm (https://github.com/vllm-project/vllm) to run it on a Couple of A100's, and you can benchmark this using this library (https://github.com/EleutherAI/lm-evaluation-harness)
- FLaNK AI for 11 March 2024
-
Show HN: We got fine-tuning Mistral-7B to not suck
Great question! scheduling workloads onto GPUs in a way where VRAM is being utilised efficiently was quite the challenge.
What we found was the IO latency for loading model weights into VRAM will kill responsiveness if you don't "re-use" sessions (i.e. where the model weights remain loaded and you run multiple inference sessions over the same loaded weights).
Obviously projects like https://github.com/vllm-project/vllm exist but we needed to build out a scheduler that can run a fleet of GPUs for a matrix of text/image vs inference/finetune sessions.
disclaimer: I work on Helix
-
Mistral CEO confirms 'leak' of new open source AI model nearing GPT4 performance
FYI, vLLM also just added experiment multi-lora support: https://github.com/vllm-project/vllm/releases/tag/v0.3.0
Also check out the new prefix caching, I see huge potential for batch processing purposes there!
- VLLM Sacrifices Accuracy for Speed
- Easy, fast, and cheap LLM serving for everyone
- vllm
- Mixtral Expert Parallelism
- Mixtral 8x7B Support
What are some alternatives?
axolotl - Go ahead and axolotl questions
TensorRT - NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source components of TensorRT.
OpenPipe - Turn expensive prompts into cheap fine-tuned models
CTranslate2 - Fast inference engine for Transformer models
Llama-2-Onnx
lmdeploy - LMDeploy is a toolkit for compressing, deploying, and serving LLMs.
trieve - All-in-one infrastructure for building search, recommendations, and RAG. Trieve combines search language models with tools for tuning ranking and relevance.
open_model_zoo - Pre-trained Deep Learning models and demos (high quality and extremely fast)
tritony - Tiny configuration for Triton Inference Server
deepeval - The LLM Evaluation Framework
faster-whisper - Faster Whisper transcription with CTranslate2