Mustafa Tahir KANAT

Rodeszones

10 21 173

AI & ML interests

None yet

Recent Activity

upvoted an article about 13 hours ago

LFM2.5 Q4\_0 Checkpoints from Quantization-Aware Distillation

commentedon an article 1 day ago

Extreme Overtraining in Tiny Language Models

commentedon an article 2 days ago

Sleeper Agents and How to Tame Them

View all activity

Organizations

None yet

upvoted an article about 13 hours ago

Article

LFM2.5 Q4\_0 Checkpoints from Quantization-Aware Distillation

LiquidAI

•

about 19 hours ago

• 25

commented on Extreme Overtraining in Tiny Language Models 1 day ago

Have you experimented with or tested different vocabulary sizes to see if the vocab size directly impacts the overtraining threshold and benchmark peak?

commented on Sleeper Agents and How to Tame Them 2 days ago

Execute Order 66

liked 3 datasets 4 days ago

upvoted 3 papers 4 days ago

On-Policy Self-Distillation without Any Supervision

Paper • 2608.06296 • Published 11 days ago • 214

Weak-to-Strong Generalization via Direct On-Policy Distillation

Paper • 2607.05394 • Published Jul 8 • 147

Weak-to-Strong On-Policy Distillation

Paper • 2607.26246 • Published 23 days ago • 57

upvoted an article 5 days ago

Article

Distillation in 2026 (so far): which frontier models use it and how

sergiopaniego

•

Jul 8

• 21

upvoted an article 9 days ago

Article

Making Knowledge Distillation Cheap Enough to Run at Scale

MultiverseComputingCAI

•

10 days ago

• 32

liked a dataset 15 days ago

HuggingFaceH4/ultrachat_200k

Viewer • Updated Oct 16, 2024 • 515k • 90.2k • 875

liked a dataset 18 days ago

Qyrou/reasoning-corpus-4K-5M-v1

Viewer • Updated 20 days ago • 3.67M • 14.2k • 204

liked 2 models 20 days ago

deepseek-ai/DeepSeek-V4-Flash-0731

Text Generation • 304B • Updated 19 days ago • 2.33M • • 3.56k

LiquidAI/LFM2.5-Encoder-350M

Fill-Mask • 0.4B • Updated 23 days ago • 8.85k • 134

liked a dataset 20 days ago

HuggingFaceCode/stack-v3-train

Preview • Updated 3 days ago • 264k • 360

liked a model 23 days ago

unsloth/gemma-4-12B-it-qat-GGUF

Any-to-Any • 12B • Updated Jul 17 • 561k • 430

liked a model 30 days ago

openbmb/MiniCPM-RobotManip

Robotics • 2B • Updated 29 days ago • 1.13k • 182

reacted to danielhanchen's post with 👍 about 1 month ago

Post

6214

Gemma 4 is now faster and much more accurate! 🚀

Google made huge improvements to tool-calling and chat accuracy, reliability + speed.
To get fixes, re-download our updated GGUF, MLX, NVFP4 quants!

Unsloth quants: https://huggingface.co/collections/unsloth/gemma-4
Gemma 4 Guide: https://unsloth.ai/docs/models/gemma-4