Donut Earthers 🍩

community

Activity Feed Request to join this org

AI & ML interests

inviting members to join the donut earth community from all over the donut 🍩

Recent Activity

p3nGu1nZz new activity about 2 months ago

donut-earth/donut-AE:enhance-setup-script

aldigobbler new activity about 2 months ago

donut-earth/donut-AE:enhance-setup-script

p3nGu1nZz updated a model about 2 months ago

donut-earth/donut-AE

View all activity

donut-earth's activity

not-lain

posted an update about 5 hours ago

Post

181

I have just released a new blogpost about kv caching and its role in inference speedup 🚀
🔗 https://huggingface.co/blog/not-lain/kv-caching/
some takeaways :

not-lain

posted an update 13 days ago

Post

1155

we now have more than 2000 public AI models using ModelHubMixin🤗

not-lain

posted an update 18 days ago

Post

3928

Published a new blogpost 📖
In this blogpost I have gone through the transformers' architecture emphasizing how shapes propagate throughout each layer.
🔗 https://huggingface.co/blog/not-lain/tensor-dims
some interesting takeaways :

alielfilali01

posted an update 23 days ago

Post

1887

3C3H AraGen Leaderboard welcomes today deepseek-ai/DeepSeek-V3 and 12 other models (including the late gpt-3.5 💀) to the ranking of best LLMs in Arabic !

Observations:
- DeepSeek-v3 ranked 3rd and only Open model among the top 5 !

- A 14B open model ( Qwen/Qwen2.5-14B-Instruct) outperforms gpt-3.5-turbo-0125 (from last year). This shows how much we came in advancing and supporting Arabic presence within the LLM ecosystem !

- Contrary to what observed in likelihood-acc leaderboards (like OALL/Open-Arabic-LLM-Leaderboard) further finetuned models like maldv/Qwentile2.5-32B-Instruct actually decreased the performance compared to the original model Qwen/Qwen2.5-32B-Instruct.
It's worth to note that the decrease is statiscally insignificant which imply that at best, the out-domain finetuning do not really hurts the model original capabilities acquired during pretraining.
Previous work addressed this (finetuning VS pretraining) but more investigation in this regard is required (any PhDs here ? This could be your question ...)

Check out the latest rankings: inceptionai/AraGen-Leaderboard

alielfilali01

posted an update about 1 month ago

Post

1916

~75% on the challenging GPQA with only 40M parameters 🔥🥳

GREAT ACHIEVEMENT ! Or is it ?

This new Work, "Data Laundering: Artificially Boosting Benchmark Results through Knowledge Distillation", take out the mystery about many models i personally suspected their results. Speacially on leaderboards other than the english one, Like the Open Arabic LLM Leaderbaord OALL/Open-Arabic-LLM-Leaderboard.

The authors of this work, first started by training a model on the GPQA data, which, unsurprisingly, led to the model achieving 100% performance.

Afterward, they trained what they referred to as a 'legitimate' model on legitimate data (MedMCQA). However, they introduced a distillation loss from the earlier, 'cheated' model.

What they discovered was fascinating: the knowledge of GPQA leaked through this distillation loss, even though the legitimate model was never explicitly trained on GPQA during this stage.

This raises important questions about the careful use of distillation in model training, especially when the training data is opaque. As they demonstrated, it’s apparently possible to (intentionally or unintentionally) leak test data through this method.

Find out more: Data Laundering: Artificially Boosting Benchmark Results through Knowledge Distillation (2412.15255)

1 reply

takarajordan

posted an update about 1 month ago

Post

1289

I made an RSS feed for HuggingFace Daily Papers!! 🤗

Just Subscribe here: https://papers.takara.ai/api/feed

It updates every 24 hours, completely written as a serverless go script with a Redis cache (to avoid hitting HF all the time).

I'm open sourcing the code, you can check out my repo and deploy it on Vercel extremely easily!
https://github.com/404missinglink/HF-Daily-Papers-Feeds

thanks to @John6666 @p3nGu1nZz for your early support

alielfilali01

posted an update about 2 months ago

Post

3453

Unpopular opinion: Open Source takes courage to do !

Not everyone is brave enough to release what they have done (the way they've done it) to the wild to be judged !
It really requires a high level of "knowing wth are you doing" ! It's kind of a super power !

Cheers to the heroes here who see this!

5 replies

takarajordan

posted an update about 2 months ago

Post

2339

I'm super excited to release my first open-source text dataset:

WorldScenario 20K is a novel dataset of 20,000 synthetically generated multi-stakeholder scenarios designed to simulate real-world decision-making processes. Each scenario explores a unique environmental, societal, or economic issue.

I used the brand new meta-llama/Llama-3.3-70B-Instruct model to generate this dataset and I put the dataset through some post processing to clean and evaluate the dataset for diversity.

I'd appreciate some feedback and thoughts on my new release! Thanks!

takarajordan/WorldScenario_20K

8 replies

alielfilali01

posted an update about 2 months ago

Post

1535

Apparently i forgot to put this here !

Well, this is a bit late but consider given our recent blog a read if you are interested in Evaluation.

You don't have to be into Arabic NLP in order to read it, the main contribution we are introducing is a new evaluation measure for NLG. We made the fisrt application of this measure on Arabic for now and we will be working with colleagues from the community to expand it to other languages.

Blog:
Rethinking LLM Evaluation with 3C3H: AraGen Benchmark and Leaderboard
https://huggingface.co/blog/leaderboard-3c3h-aragen

Space:
inceptionai/AraGen-Leaderboard

Give it a read and let me know your thoughts 🤗

p3nGu1nZz

in donut-earth/donut-AE about 2 months ago

enhance-setup-script

#1 opened about 2 months ago by

p3nGu1nZz

aldigobbler

in donut-earth/donut-AE about 2 months ago

enhance-setup-script

#1 opened about 2 months ago by

p3nGu1nZz

updated a model about 2 months ago

donut-earth/donut-AE

Updated Dec 1, 2024 • 3

aldigobbler

updated a Space 2 months ago

Running

🚀

README

aldigobbler

updated a model 2 months ago

donut-earth/donut-AE

Updated Dec 1, 2024 • 3

takarajordan

posted an update 2 months ago

Post

1210

I'm not sure why I haven't done this already!

I just made a space to count and visualize tokens for Diffusion models, no more guesswork! It's super fast too.

Check it out here and try out your prompts: takarajordan/DiffusionTokenizer

Uses these tokenizers below:
openai/clip-vit-large-patch14
google/t5-v1_1-xxl

aldigobbler

posted an update 2 months ago

Post

1604

p104-100s are beasts. 8 gigs of VRAM, 12 tok/s on qwen 14b at q4, and 18 tok/s on 7b at q6. best thing - 20 euros each.

https://furry.engineer/@cappuch/113500349547803802