Ample Project
🔥
AI & ML interests
Gradio, AI Agents, MCP
Recent Activity
View all activity
noumanjavaid
published
a
Space
about 7 hours ago
Post
508
I've uploaded abliteration code with support for sparsification of the refusal vector. It's poorly documented, but the code should be straightforward.
https://github.com/jim-plus/llm-abliteration
The code is built atop a fork that enabled abliteration to be performed on models loaded in 4-bit or 8-bit bitsandbytes quantization. TransformerLens is not required, just plain Transformers. For those previously unaware, this opens up abliteration experimentation to more people with local VRAM limitations.
Since performing abliteration on a quant involves precision and perplexity loss, it stands to reason that a small amount of magnitude sparsification could filter out some noise and possibly even reduce the damage inflicted on latent space via ablation of the refusal vector.
There's a small but real acceleration of ablation of the refusal vector by reducing outer product operations from O(d²×n) to O(d×n), and then by pushing said computation layerwise to GPU. The code is hardcoded for CUDA acceleration currently. Normalization of the refusal vector was deferred in order to allow sparsification. In principle other behavior vector interventions could also be added and explored.
https://github.com/jim-plus/llm-abliteration
The code is built atop a fork that enabled abliteration to be performed on models loaded in 4-bit or 8-bit bitsandbytes quantization. TransformerLens is not required, just plain Transformers. For those previously unaware, this opens up abliteration experimentation to more people with local VRAM limitations.
Since performing abliteration on a quant involves precision and perplexity loss, it stands to reason that a small amount of magnitude sparsification could filter out some noise and possibly even reduce the damage inflicted on latent space via ablation of the refusal vector.
There's a small but real acceleration of ablation of the refusal vector by reducing outer product operations from O(d²×n) to O(d×n), and then by pushing said computation layerwise to GPU. The code is hardcoded for CUDA acceleration currently. Normalization of the refusal vector was deferred in order to allow sparsification. In principle other behavior vector interventions could also be added and explored.
Post
1017
What other features would you like to see on the Trackio Dashboard? (
gradio-templates/trackio-dashboard)
aldigobbler
posted
an
update
2 months ago
Post
338
hop on huggingface minecraft server
aldigobbler/mc-server
1.20.1 (view info on space)
soon to add agents into the game
aldigobbler/mc-server
1.20.1 (view info on space)
soon to add agents into the game
lukmanaj
authored
3
papers
3 months ago
Who Wrote This? Identifying Machine vs Human-Generated Text in Hausa
Paper
•
2503.13101
•
Published
•
1
Leveraging Closed-Access Multilingual Embedding for Automatic Sentence Alignment in Low Resource Languages
Paper
•
2311.12179
•
Published
Evaluating Deep Learning Models for African Wildlife Image Classification: From DenseNet to Vision Transformers
Paper
•
2507.21364
•
Published
•
4
elismasilva
posted
an
update
3 months ago
Post
347
Building settings panels in? I created "PropertySheet", a new gradio component that turns your Python dataclass into a full UI! ⚡️
✅ Sliders, dropdowns, color pickers & more
✅ Collapsible groups
✅ Reset buttons & tooltips
elismasilva/gradio_propertysheet
✅ Sliders, dropdowns, color pickers & more
✅ Collapsible groups
✅ Reset buttons & tooltips
elismasilva/gradio_propertysheet
elismasilva
posted
an
update
4 months ago
Post
304
hi guys i created this gradio component based on Xenova's tokenizer playground application. This will help me in an application I'm building, I hope it will be useful for you too.
elismasilva/gradio_tokenizertextbox
gradio_tokenizertextbox - a Hugging Face Space by elismasilva
elismasilva/gradio_tokenizertextbox
gradio_tokenizertextbox - a Hugging Face Space by elismasilva
abdur75648
authored
3
papers
4 months ago
UTRNet: High-Resolution Urdu Text Recognition In Printed Documents
Paper
•
2306.15782
•
Published
•
7
V-Zen: Efficient GUI Understanding and Precise Grounding With A Novel Multimodal LLM
Paper
•
2405.15341
•
Published
•
7
Echo-DND: A dual noise diffusion model for robust and precise left ventricle segmentation in echocardiography
Paper
•
2506.15166
•
Published
•
7
Post
327
Having an incredible time at the Agents MCP Hackathon
The learning curve has been steep but absolutely rewarding - diving deep into Model Context Protocol (MCP), building intelligent agents, and exploring the latest in LLM capabilities. There's something magical about seeing these technologies come together in real-time applications.
Huge shoutout to the hackathon organizers and sponsors for providing generous API credits from top inference providers. It's removed all the barriers to experimentation and let us focus purely on innovation and creativity. This level of support really makes the difference between having an idea and actually bringing it to life.
The community energy here is infectious - collaborating with brilliant minds, sharing breakthroughs, and learning from each other's approaches. Whether you're debugging MCP connections at 2 AM or celebrating a successful agent deployment, there's always someone ready to help or celebrate with you.
Interested in joining or learning more? Check out the hackathon details here:
Agents-MCP-Hackathon
What a time to be building in AI! 🔥
The learning curve has been steep but absolutely rewarding - diving deep into Model Context Protocol (MCP), building intelligent agents, and exploring the latest in LLM capabilities. There's something magical about seeing these technologies come together in real-time applications.
Huge shoutout to the hackathon organizers and sponsors for providing generous API credits from top inference providers. It's removed all the barriers to experimentation and let us focus purely on innovation and creativity. This level of support really makes the difference between having an idea and actually bringing it to life.
The community energy here is infectious - collaborating with brilliant minds, sharing breakthroughs, and learning from each other's approaches. Whether you're debugging MCP connections at 2 AM or celebrating a successful agent deployment, there's always someone ready to help or celebrate with you.
Interested in joining or learning more? Check out the hackathon details here:
Agents-MCP-Hackathon
What a time to be building in AI! 🔥
Post
4425
The Gradio x Agents x MCP hackathon keeps growing! We now have more $1,000,000 in credit for participants and and >$16,000 in cash prizes for winners.
We've kept registration open until the end of this week, so join and let's build cool stuff together as a community: https://huggingface.co/spaces/ysharma/gradio-hackathon-registration-2025
We've kept registration open until the end of this week, so join and let's build cool stuff together as a community: https://huggingface.co/spaces/ysharma/gradio-hackathon-registration-2025
aldigobbler
posted
an
update
5 months ago
Post
356
no ai slop posted here today i just feel like posting what i did for today
wrote a little framework for turning multiple dense models (llama based) into Sparse MoEs.. i found it fun, spent the whole day and a half on it.
code @ https://gist.github.com/cappuch/6a454ec8d2d349a27f9fd84f6ac90554
wrote a little framework for turning multiple dense models (llama based) into Sparse MoEs.. i found it fun, spent the whole day and a half on it.
code @ https://gist.github.com/cappuch/6a454ec8d2d349a27f9fd84f6ac90554
Post
2455
I am so happy to share to all that I’ve just completed the first unit of the new MCP course on Hugging Face and earned my certificate! The AI acceleration track is intense and fast-paced, but I’m doing my best to keep up. Excited for what’s ahead!
- 1 reply
Post
5195
HOW TO ADD MCP SUPPORT TO ANY 🤗 SPACE
Gradio now supports MCP! If you want to convert an existing Space, like this one hexgrad/Kokoro-TTS, so that you can use it with Claude Desktop / Cursor / Cline / TinyAgents / or any LLM that supports MCP, here's all you need to do:
1. Duplicate the Space (in the Settings Tab)
2. Upgrade the Gradio
3. Set
4. (Optionally) add docstrings to the function so that the LLM knows how to use it, like this:
That's it! Now your LLM will be able to talk to you 🤯
Gradio now supports MCP! If you want to convert an existing Space, like this one hexgrad/Kokoro-TTS, so that you can use it with Claude Desktop / Cursor / Cline / TinyAgents / or any LLM that supports MCP, here's all you need to do:
1. Duplicate the Space (in the Settings Tab)
2. Upgrade the Gradio
sdk_version to 5.28 (in the README.md)3. Set
mcp_server=True in launch()4. (Optionally) add docstrings to the function so that the LLM knows how to use it, like this:
def generate(text, speed=1):
"""
Convert text to speech audio.
Parameters:
text (str): The input text to be converted to speech.
speed (float, optional): Playback speed of the generated speech.That's it! Now your LLM will be able to talk to you 🤯
Post
2263
I’m excited to share that I’ve completed the Hugging Face Agents Course and earned my certificate.
Over the past few months, I explored how to build intelligent, autonomous agents using cutting-edge tools like smolagents, LlamaIndex, and LangGraph. The course covered everything from the fundamentals of agents to advanced topics like fine-tuning for function-calling, observability, evaluation, and even agents in games.
Some key content included:
1. Introduction to AI Agents
2. Agentic RAG use cases
3. Multi-framework implementation: smolagents, LlamaIndex, and LangGraph
4. Building, testing, and certifying a complete agent project
This was a hands-on, practical experience that deepened my understanding of how to design reliable, tool-using LLM agents. Looking forward to leveraging these skills in real-world applications in healthcare, logistics, and beyond.
Many thanks to the Hugging Face team for putting this together.
Let’s build safe and useful agents!
Over the past few months, I explored how to build intelligent, autonomous agents using cutting-edge tools like smolagents, LlamaIndex, and LangGraph. The course covered everything from the fundamentals of agents to advanced topics like fine-tuning for function-calling, observability, evaluation, and even agents in games.
Some key content included:
1. Introduction to AI Agents
2. Agentic RAG use cases
3. Multi-framework implementation: smolagents, LlamaIndex, and LangGraph
4. Building, testing, and certifying a complete agent project
This was a hands-on, practical experience that deepened my understanding of how to design reliable, tool-using LLM agents. Looking forward to leveraging these skills in real-world applications in healthcare, logistics, and beyond.
Many thanks to the Hugging Face team for putting this together.
Let’s build safe and useful agents!
-
- 7 replies