community-events
sentencepiece
community-events | sentencepiece | |
---|---|---|
8 | 19 | |
379 | 9,520 | |
2.1% | 2.1% | |
7.2 | 8.1 | |
5 months ago | 3 days ago | |
Jupyter Notebook | C++ | |
- | 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.
community-events
-
Controlling Stable Diffusion with JAX & Diffusers using TPU v4
Best applications that will come out of this sprint will receive prizes. You can find more information here. If you want to get started, simply join huggingface.co/discord, take the role 🧨 Diffusers and head to #jax-diffusers-ideas to share your idea or join one of the teams, and fill this form: https://forms.gle/t3M7aNPuLL9V1sfa9
-
JAX & Diffusers to Control Stable Diffusion (with TPUs ⚡️ )
It will start on 17th of April. To join us, you can join huggingface.co/join/discord and take the role Diffusers from #role-assignment. After this, simply fill the form provided in this guide to later get access to TPUs. https://github.com/huggingface/community-events/tree/main/jax-controlnet-sprint
- “Control Stable Diffusion” Sprint kicks off with free TPU-v4 from Google
-
Free compute to train custom ControlNet by Hugging Face
Details and sign-up: https://github.com/huggingface/community-events/tree/main/jax-controlnet-sprint
-
How can I create a dataset to refine Whisper AI from old videos with subtitles?
For the training, I extremely recommend checking out the Whisper Fine-Tuning Event. It has a python script to train in one command, tons of tips, even a walkthrough video.
- I am using OpenAi's whisper transcription/translation model. I am wondering if I can improve it's performance by optimizing the audio files somehow. What features of audio files should I look into to make the whisper model perform better?
-
[N] Gradio Blocks + Hugging Face event is starting this week. A hackathon type event from May 17th to May 31st with prizes in which we will create interactive web demos for state-of-the-art machine learning models
We are happy to invite you to the Gradio Blocks Party - a community event in which we will create interactive demos for state-of-the-art machine learning models. Demos are powerful because they allow anyone — not just ML engineers — to try out models in the browser, give feedback on predictions, identify trustworthy models. The event will take place from May 17th to 31st. We will be organizing this event on Github and the Hugging Face discord channel. Prizes will be given at the end of the event, see the Prizes section
-
Dall-E 2
If you're interested in generative models, Hugging Face is putting on an event around generative models right now called the HugGAN sprint, where they're giving away free access to compute to train models like this.
You can join it by following the steps in the guide here: https://github.com/huggingface/community-events/tree/main/hu...
There will also be talks from awesome folks at EleutherAI, Google, and Deepmind
sentencepiece
- sentencepiece
-
LLM.int8(): 8-Bit Matrix Multiplication for Transformers at Scale
you need to train the model on 1 trillion tokens (https://platform.openai.com/tokenizer https://github.com/google/sentencepiece) anyways for it to get reasoning capacities, which it feels very unlikely that your data is that much.
I'm highly skeptical that you have enough data to pretrain if you don't have enough data to fine tune.
fine tuning + vector search + prompting of as much stuff as you can, on a LLM like palm2 or gpt4 is what I would do. otherwise you can use falcon 40B ofc.
maybe I should charge for this ahah
-
[P] TokenMonster Ungreedy ~ 35% faster inference and 35% increased context-length for large language models (compared to tiktoken). Benchmarks included.
a) Comparison with SentencePiece tokenizer with comparable settings (It can also ignore word-boundaries and create phrase tokens)
-
LLaMA tokenizer: is a JavaScript implementation available anywhere?
LLaMA uses the sentencepiece tokenizer: https://github.com/google/sentencepiece
-
[P] New tokenization method improves LLM performance & context-length by 25%+
Besides, are you familiar with SentencePiece? What you are doing looks very similar (generate a large vocab, prune worst token until vocab size is reached), only the token selection criterion is different. It's also purely data driven in the sense that there are no assumption specific to language (and it can optionally segment across whitespace, as you are doing).
-
Code runs without definition of function (automatically calls a different function instead)
Hi, I'm studying the implementation of encode and decode functions for Google's SentencePiece tokenizer.
-
How to handle multiple languages in a sentence?
I think many LMs nowadays use unicode tokenizers, that are not tied to specific languages. E.g. sentencepiece is the most popular one: https://github.com/google/sentencepiece
- Large language models are having their Stable Diffusion moment
-
LLaMA-7B in Pure C++ with full Apple Silicon support
If you are interested in implementing LLaMA yourself or learning, I noticed that the reference code by Facebook is one of the cleaner, easier to read ML code I've seen in a while. https://github.com/facebookresearch/llama/blob/main/llama/mo... It's about 200 lines long. You probably do need a bit of knowledge to understand what you are reading but I was pleasantly surprised.
For example in comparison, StableDiffusion torch code in diffusers and transformers Python libraries has lots of conditionals, experiments etc. that are not being used that can make it hard to follow what is going on.
Last weekend I got the "main loop" of the transformer working in pure CPU Rust code, following the reference code. My crappy code is just very very slow as I focused on getting it to run, not making it fast. The tokenizer uses some Google thing https://github.com/google/sentencepiece but luckily for inference it seems that you just need to be able to parse the tokenizer model file and not understand how it was created; I was able to strip out the protobuf files from that repository and add it to Rust and read the tokens.
I am optimistic that someone makes a high quality CPU or some CPU+GPU+SSD combination thingmaling that will make it somewhat practical to run even the large LLM models without needing an A100 or two.
- ChatGPT in an iOS Shortcut – Worlds Smartest HomeKit Voice Assistant
What are some alternatives?
dalle-2-preview
transformers - 🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
dalle-mini - DALL·E Mini - Generate images from a text prompt
CTranslate2 - Fast inference engine for Transformer models
bevy_retro - Plugin pack for making 2D games with Bevy
llama - Inference code for Llama models
lm-human-preferences - Code for the paper Fine-Tuning Language Models from Human Preferences
gpt-2 - Code for the paper "Language Models are Unsupervised Multitask Learners"
gpt-3 - GPT-3: Language Models are Few-Shot Learners
text-generation-webui - A Gradio web UI for Large Language Models. Supports transformers, GPTQ, AWQ, EXL2, llama.cpp (GGUF), Llama models.
glide-text2im - GLIDE: a diffusion-based text-conditional image synthesis model
OpenNMT-Tutorial - Neural Machine Translation (NMT) tutorial. Data preprocessing, model training, evaluation, and deployment.