But higher LR makes the path more volatile/unstable ? @nroggendorff
Hence the larger wheels, making it more difficult to enter holes at high speed.
Noa Roggendorff
nroggendorff
AI & ML interests
None defined yet.
Recent Activity
Organizations
reacted to
AdinaY's
post with 👍
3 days ago
Post
2564
From ChatGPT Healthcare to Claude for healthcare, AI in medicine is speeding up🚀
Now BaichuanAI joins with Baichuan-M3 🏥 an open medical LLM trained for clinical decision-making
https://huggingface.co/collections/baichuan-inc/baichuan-m3
✨ 235B - Apache2.0
✨ Lower hallucinations via Fact-Aware RL
✨ Built for long medical chats
Now BaichuanAI joins with Baichuan-M3 🏥 an open medical LLM trained for clinical decision-making
https://huggingface.co/collections/baichuan-inc/baichuan-m3
✨ 235B - Apache2.0
✨ Lower hallucinations via Fact-Aware RL
✨ Built for long medical chats
-
- 2 replies
Thanks for including human performance. I was curious about that.
reacted to
Bochkov's
post with 👀
3 days ago
Post
1915
Curious reproducible fact: I trained a GPT-like decoder-only Transformer where the entire input embedding table is frozen and reduced to a 16‑D binary token-ID code (0/1) — this is NOT 16-bit quantization.
Key details:
- vocab_size = 65536, n_embed = 16 (2^16 = 65536 unique IDs)
- deterministic expansion 16 → d_model=1024 via repeat_interleave (scale=64)
- full embedding table is published (embeddings.txt) for auditability
Repro note + verification script:
https://huggingface.co/blog/Bochkov/emergent-semantics-beyond-token-embeddings
Model repo:
Bochkov/emergent-semantics-model-16-bit-269m
License: Apache-2.0
Key details:
- vocab_size = 65536, n_embed = 16 (2^16 = 65536 unique IDs)
- deterministic expansion 16 → d_model=1024 via repeat_interleave (scale=64)
- full embedding table is published (embeddings.txt) for auditability
Repro note + verification script:
https://huggingface.co/blog/Bochkov/emergent-semantics-beyond-token-embeddings
Model repo:
Bochkov/emergent-semantics-model-16-bit-269m
License: Apache-2.0
reacted to
prithivMLmods's
post with 👍
3 days ago
Post
4144
Update: TRELLIS.2 (Text to 3D, Image to 3D) Gradio with Rerun Embedded demo with improved visualization of the 3D model previewer is now available on Hugging Face. Generate assets and view them in the 3D viewer, powered and streamlined with Microsoft’s TRELLIS.2 and Tongyi-MAI’s Z-Image-Turbo models.
🤗 TRELLIS.2 (Demo): prithivMLmods/TRELLIS.2-Text-to-3D
🕹️ GitHub: https://github.com/PRITHIVSAKTHIUR/TRELLIS.2-Text-to-3D-RERUN
🕹️ Collection: https://huggingface.co/collections/prithivMLmods/multimodal-implementations
To know more about it, visit the app page or the respective model page!
🤗 TRELLIS.2 (Demo): prithivMLmods/TRELLIS.2-Text-to-3D
🕹️ GitHub: https://github.com/PRITHIVSAKTHIUR/TRELLIS.2-Text-to-3D-RERUN
🕹️ Collection: https://huggingface.co/collections/prithivMLmods/multimodal-implementations
To know more about it, visit the app page or the respective model page!
reacted to
Ujjwal-Tyagi's
post with 🤗
3 days ago
Post
2508
I am very excited to see the release of
nyuuzyou/gitee-code. This is exactly what I have been looking for. Thank you to
@nyuuzyou
for his hard work on this.
- 3 replies
Batch size is the number of holes, or the number of cars. Learning rate is your wheel size (assuming you have enough torque to make it so having a bigger wheel doesn't decrease your wheel RPM).
reacted to
prithivMLmods's
post with 🤗
17 days ago
Post
4213
Introducing the Qwen-Image-Edit-2511-LoRAs-Fast demo, featuring image property comparison and contrast, built on top of Gradio and the combined Rerun SDK. It supports single and multi-image edits with existing LoRAs that are lazily loaded. (Note: This is still an experimental Space for Qwen-Image-Edit-2511.)
⭐ Space Demo: prithivMLmods/Qwen-Image-Edit-2511-LoRAs-Fast
⭐ GitHub: https://github.com/PRITHIVSAKTHIUR/Qwen-Image-Edit-2511-LoRAs-Fast-Multi-Image-Rerun
⭐ Collection: https://huggingface.co/collections/prithivMLmods/image-generation-apps-collection
To know more about it, visit the app page or the respective model page!
⭐ Space Demo: prithivMLmods/Qwen-Image-Edit-2511-LoRAs-Fast
⭐ GitHub: https://github.com/PRITHIVSAKTHIUR/Qwen-Image-Edit-2511-LoRAs-Fast-Multi-Image-Rerun
⭐ Collection: https://huggingface.co/collections/prithivMLmods/image-generation-apps-collection
To know more about it, visit the app page or the respective model page!
-
- 2 replies
replied to
prithivMLmods's
post
17 days ago
very cool
replied to
their
post
about 2 months ago
I'm baffled by the silence from staff regarding this...
They aren't silent, they're just really quiet. You have to look for them. Not that that's how it should be, of course.
replied to
their
post
about 2 months ago
Thanks for being so helpful.
replied to
their
post
about 2 months ago
Don't "Lol" my peril! /j
replied to
their
post
about 2 months ago
It's even worse because sometimes the spaces fail to schedule for no reason. Sometimes it's because they don't have enough hardware capacity, but sometimes it doesn't even give a reason. It just says "failed to schedule". I would not want to pay for such an inconsistent service anyway.
replied to
their
post
about 2 months ago
This comment has been hidden
replied to
their
post
about 2 months ago
all my spaces are paused and/or on cpu (free) or zerogpu
the paused spaces are on cpu (free)
reacted to
their
post with ❤️
about 2 months ago
Post
2901
I am now being charged for paused and unstarted spaces out of the blue.
I think this is it, folks. o7
The unstarted spaces I can get behind. I would've appreciated a warning email first, but whatever. However, every time I restart the active usage goes up, despite all of my spaces being moved to CPU (free), and being paused.
I think this is it, folks. o7
The unstarted spaces I can get behind. I would've appreciated a warning email first, but whatever. However, every time I restart the active usage goes up, despite all of my spaces being moved to CPU (free), and being paused.
-
- 13 replies
posted
an
update
about 2 months ago
Post
2901
I am now being charged for paused and unstarted spaces out of the blue.
I think this is it, folks. o7
The unstarted spaces I can get behind. I would've appreciated a warning email first, but whatever. However, every time I restart the active usage goes up, despite all of my spaces being moved to CPU (free), and being paused.
I think this is it, folks. o7
The unstarted spaces I can get behind. I would've appreciated a warning email first, but whatever. However, every time I restart the active usage goes up, despite all of my spaces being moved to CPU (free), and being paused.
-
-
-
- 13 replies
reacted to
their
post with 😔
about 2 months ago
Post
2375
Developing with ZeroGPU without a PRO account is painful. They give you so many requests at once, but then have like a 24 hour cooldown. I vote less requests in a batch, but then a shorter cooldown.
or just less of a cooldown, but i understand if that is not allowed
or just less of a cooldown, but i understand if that is not allowed
-
-
- 3 replies
replied to
their
post
2 months ago
Why don't you get pro, pep?
posted
an
update
2 months ago
Post
2375
Developing with ZeroGPU without a PRO account is painful. They give you so many requests at once, but then have like a 24 hour cooldown. I vote less requests in a batch, but then a shorter cooldown.
or just less of a cooldown, but i understand if that is not allowed
or just less of a cooldown, but i understand if that is not allowed
-
-
- 3 replies