onnx-community/Carbon-500M-ONNX
Text Generation • Updated • 34 • 2
Pulmo - The AI Radiology Assistant PRO
pulmo
AI & ML interests
CXR's, Biology, ML, and Open Source Software
Recent Activity
liked a model 3 days ago
onnx-community/Carbon-500M-ONNX upvoted a paper 3 days ago
optimize_anything: A Universal API for Optimizing any Text Parameter reacted to salma-remyx's post with 👍 3 days ago
Just trained a 2B coding model to rank candidate AI/ML research ideas against the implicit preferences in a code repository's merge history.
The training data comes from a Gaussian Process fit on the accumulated dispositions in VQASynth, where each PR against a deployed project yields a pairwise comparison between the feature branch preferred and the baseline at main.
The GP scores candidate papers to synthesize preference pairs, and DPO with LoRA bakes the ranking pipeline into the model's weights.
After 1 epoch the model reaches 87.4% reward accuracy on the held-out eval split against 92.3% on training, consistent with learning the task without overfitting.
Now, I'm scaling the pipeline to thousands of repos for a generalization test.
Dataset: https://huggingface.co/datasets/remyxai/mhpd-dpo-v0
Model: https://huggingface.co/remyxai/mhpd-dpo-qwen3.5-2b-vqasynth
Substack: https://remyxai.substack.com/p/the-ai-pmOrganizations
liked a
model 3 days ago
upvoted a paper 3 days ago
reacted to salma-remyx's post with 👍 3 days ago
Post
5057
Just trained a 2B coding model to rank candidate AI/ML research ideas against the implicit preferences in a code repository's merge history.
The training data comes from a Gaussian Process fit on the accumulated dispositions in VQASynth, where each PR against a deployed project yields a pairwise comparison between the feature branch preferred and the baseline at main.
The GP scores candidate papers to synthesize preference pairs, and DPO with LoRA bakes the ranking pipeline into the model's weights.
After 1 epoch the model reaches 87.4% reward accuracy on the held-out eval split against 92.3% on training, consistent with learning the task without overfitting.
Now, I'm scaling the pipeline to thousands of repos for a generalization test.
Dataset: remyxai/mhpd-dpo-v0
Model: remyxai/mhpd-dpo-qwen3.5-2b-vqasynth
Substack: https://remyxai.substack.com/p/the-ai-pm
The training data comes from a Gaussian Process fit on the accumulated dispositions in VQASynth, where each PR against a deployed project yields a pairwise comparison between the feature branch preferred and the baseline at main.
The GP scores candidate papers to synthesize preference pairs, and DPO with LoRA bakes the ranking pipeline into the model's weights.
After 1 epoch the model reaches 87.4% reward accuracy on the held-out eval split against 92.3% on training, consistent with learning the task without overfitting.
Now, I'm scaling the pipeline to thousands of repos for a generalization test.
Dataset: remyxai/mhpd-dpo-v0
Model: remyxai/mhpd-dpo-qwen3.5-2b-vqasynth
Substack: https://remyxai.substack.com/p/the-ai-pm
liked 2
models 3 days ago
liked a
Space 3 days ago
updated a
dataset 11 days ago
liked a
Space 11 days ago
reacted to danielhanchen's post with 🚀 11 days ago
Post
5707
We’re excited to announce that Unsloth has joined the PyTorch Ecosystem! 🔥🦥
Unsloth is an open-source project that makes training & running models more accurate and faster with less compute. Our mission is to make local AI accessible to everyone. Thanks to all of you for making this possible! 💕
Blog: https://unsloth.ai/blog/pytorch
GitHub: https://github.com/unslothai/unsloth
Unsloth is an open-source project that makes training & running models more accurate and faster with less compute. Our mission is to make local AI accessible to everyone. Thanks to all of you for making this possible! 💕
Blog: https://unsloth.ai/blog/pytorch
GitHub: https://github.com/unslothai/unsloth
- 2 replies
reacted to danielhanchen's post with 🔥 14 days ago
Post
7648
We collaborated with NVIDIA to teach you how we made LLM training ~25% faster! 🚀
Learn how 3 optimizations help your home GPU train models faster:
1. Packed-sequence metadata caching
2. Double-buffered checkpoint reloads
3. Faster MoE routing
Guide: https://unsloth.ai/blog/nvidia-collab
GitHub: https://github.com/unslothai/unsloth
Learn how 3 optimizations help your home GPU train models faster:
1. Packed-sequence metadata caching
2. Double-buffered checkpoint reloads
3. Faster MoE routing
Guide: https://unsloth.ai/blog/nvidia-collab
GitHub: https://github.com/unslothai/unsloth
Update README.md
#1 opened 20 days ago
by pulmo
updated a
dataset 23 days ago
published a
dataset 23 days ago
updated a
dataset 23 days ago