Jeff Cook

jeffcookio's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a collection 5 days ago

Helios

Collection

Helios: 14B Real-Time Long Video Generation Model can be Cheaper, Faster but Keep Stronger than 1.3B ones • 7 items • Updated 3 days ago • 17

liked 2 models 6 days ago

EricRollei/HunyuanImage-3.0-Instruct-INT8-v2

Text-to-Image • 83B • Updated 16 days ago • 145 • 1

tencent/HunyuanImage-3.0-Instruct

Image-to-Image • Updated Feb 3 • 2.42k • • 373

upvoted 2 articles 15 days ago

view article

Article

Text-to-image Architectural Experiments

• 

Nov 13, 2025

•

53

view article

Article

Training Design for Text-to-Image Models: Lessons from Ablations

• 

Feb 3

•

69

liked 2 models 20 days ago

OmniSVG/OmniSVG

Text Generation • Updated Jul 21, 2025 • 384 • 191

OmniSVG/OmniSVG1.1_8B

Text Generation • Updated Dec 3, 2025 • 11

liked a model 23 days ago

openbmb/MiniCPM-SALA

Text Generation • Updated 26 days ago • 5.86k • 489

updated a model 23 days ago

jeffcookio/DeepSeek-R1-Distill-Qwen-14B-gptqmodel-4b-32g

15B • Updated 23 days ago • 12

published a model 23 days ago

jeffcookio/DeepSeek-R1-Distill-Qwen-14B-gptqmodel-4b-32g

15B • Updated 23 days ago • 12

liked a model 23 days ago

FireRedTeam/FireRed-Image-Edit-1.0

Image-to-Image • Updated 23 days ago • 10.2k • • 311

reacted to MaziyarPanahi's post with 👀 28 days ago

view post

Post

1249

From Golden Gate Bridge to Broken JSON: Why Anthropic's SAE Steering Fails for Structured Output

I ran 6 experiments trying to use Anthropic's SAE steering for JSON generation.

- Base model: 86.8% valid JSON
- Steering only: 24.4%
- Fine-tuned: 96.6%
- FSM constrained: 100%

Steering is for semantics, not syntax.

https://huggingface.co/blog/MaziyarPanahi/sae-steering-json

See translation

👀

2

2

🚀

1

1

🤯

1

1

+

Reply

liked 4 models about 1 month ago

openbmb/MiniCPM-o-4_5

Any-to-Any • 9B • Updated 2 days ago • 88.4k • 903

alibaba-pai/FLUX.2-dev-Fun-Controlnet-Union

Updated 24 days ago • 5.57k • 25

stepfun-ai/Step-3.5-Flash

Text Generation • 199B • Updated about 10 hours ago • 255k • • 698

Tongyi-MAI/Z-Image

Text-to-Image • Updated Jan 28 • 99.6k • • 977

liked 3 models about 2 months ago

alibaba-pai/Qwen-Image-2512-Fun-Controlnet-Union

Updated 24 days ago • 20.5k • 43

zai-org/GLM-Image

Text-to-Image • Updated Jan 15 • 11.9k • • 1.03k

BennyDaBall/qwen3-4b-Z-Image-Engineer

4B • Updated Jan 19 • 4.5k • 169

liked a model 2 months ago

Lightricks/LTX-2

Image-to-Video • Updated 7 days ago • 1.61M • • 1.63k