GRPO checkpoints organized by staleness cap, with the shared 17,005-row DAPO math dataset, provenance and benchmark scores.
🏝️ On Vacation
Mohamad Zbib
zbeeb
AI & ML interests
KAUST - AUB
Recent Activity
updated a collection about 4 hours ago
Staleness: GRPO Models & Data updated a model about 4 hours ago
zbeeb/Qwen2.5-3B-GRPO-Staleness-8 published a model about 4 hours ago
zbeeb/Qwen2.5-3B-GRPO-Staleness-8Organizations
models 25
zbeeb/Qwen2.5-3B-GRPO-Staleness-8
Text Generation • 3B • Updated
zbeeb/Qwen2.5-Math-1.5B-GRPO-Staleness-8
Text Generation • 2B • Updated
zbeeb/Qwen2.5-3B-GRPO-Staleness-4
Text Generation • 3B • Updated
zbeeb/Qwen2.5-Math-1.5B-GRPO-Staleness-6
Text Generation • 2B • Updated
zbeeb/Qwen2.5-Math-1.5B-GRPO-Staleness-4
Text Generation • 2B • Updated • 169
zbeeb/Qwen2.5-Math-7B-GRPO-Staleness-2
Text Generation • 8B • Updated • 425
zbeeb/Qwen2.5-3B-GRPO-Staleness-2
Text Generation • 3B • Updated • 426
zbeeb/Qwen2.5-Math-1.5B-GRPO-Staleness-2
Text Generation • 2B • Updated • 411
zbeeb/deepseek-r1-distill-qwen-14b-fast-math-r1-sft-10ep
Text Generation • 841k • Updated • 16
zbeeb/Qwen2.5-0.5B-Instruct-Math-SFT-100K-2ep
Text Generation • 0.6B • Updated • 31
datasets 9
zbeeb/Staleness-GRPO-DAPO-Math-17k
Viewer • Updated • 17k • 43
zbeeb/dapo-math-17k-qwen3-clean
Viewer • Updated • 17k • 24
zbeeb/beneficial-dpo-dataset
Viewer • Updated • 10k • 30
zbeeb/TAPS-Datasets
Viewer • Updated • 210k • 36 • 4
zbeeb/mixeddatasafety7030
Viewer • Updated • 10k • 3
zbeeb/llama3_MathInstruct_data
Updated • 2
zbeeb/unsafe
Viewer • Updated • 200 • 4
zbeeb/safe
Updated • 17
zbeeb/Benign
Updated • 1