Amine Dirhoussi

aminediroHF

aminediro

AI & ML interests

None yet

Recent Activity

updated a Space 4 days ago

aminediroHF/moe-recipe-builder

updated a Space 6 days ago

aminediroHF/qwen3-sft-benchmark

updated a dataset 6 days ago

aminediroHF/moe-recipe-builder-catalog

View all activity

Organizations

updated a Space 4 days ago

MoE Recipe Builder

🧱

Tetris-style recipe builder for Qwen3-30B-A3B MoE training

updated a Space 6 days ago

Qwen3 SFT Benchmark

📊

MFU / TPS / memory across configs on H100 — transformers 5.6

updated a dataset 6 days ago

aminediroHF/moe-recipe-builder-catalog

Updated 6 days ago • 28

published a dataset 6 days ago

aminediroHF/moe-recipe-builder-catalog

Updated 6 days ago • 28

published a Space 7 days ago

MoE Recipe Builder

🧱

Tetris-style recipe builder for Qwen3-30B-A3B MoE training

updated a Space 20 days ago

Defeating the trainer-generator precision mismatch in TRL

🎯

Download research PDF (Pro access required)

published a Space 22 days ago

Qwen3 SFT Benchmark

📊

MFU / TPS / memory across configs on H100 — transformers 5.6

updated a bucket 22 days ago

aminediroHF/async_wordle_trackio-bucket

766 kB

updated a model 25 days ago

aminediroHF/Qwen3-235B-A22B-fused

Text Generation • 235B • Updated 25 days ago • 127

published a model 25 days ago

aminediroHF/Qwen3-235B-A22B-fused

Text Generation • 235B • Updated 25 days ago • 127

updated a model 27 days ago

aminediroHF/Qwen3-30B-A3B-fused

Text Generation • 31B • Updated 27 days ago • 384

published a model 27 days ago

aminediroHF/Qwen3-30B-A3B-fused

Text Generation • 31B • Updated 27 days ago • 384

updated a bucket 29 days ago

aminediroHF/async-grpo-wordle-deltas

17.2 GB

published a bucket 29 days ago

aminediroHF/async-grpo-wordle-deltas

17.2 GB

updated a Space 29 days ago

Async Wordle Trackio

🚀

Display interactive tracking dashboard

published a bucket 29 days ago

aminediroHF/async_wordle_trackio-bucket

766 kB

published a Space 29 days ago

Async Wordle Trackio

🚀

Display interactive tracking dashboard

updated a Space 29 days ago

Wordle Environment (High Capacity)

🟩

published a Space 29 days ago

Wordle Environment (High Capacity)

🟩

updated a Space about 1 month ago

Async Grpo Figures

🏢

Explore interactive charts for AsyncGRPO precision analysis

Amine Dirhoussi

AI & ML interests

Recent Activity

Organizations

aminediroHF's activity

MoE Recipe Builder

Qwen3 SFT Benchmark

MoE Recipe Builder

Defeating the trainer-generator precision mismatch in TRL

Qwen3 SFT Benchmark

aminediroHF/async_wordle_trackio-bucket

aminediroHF/async-grpo-wordle-deltas

aminediroHF/async-grpo-wordle-deltas

Async Wordle Trackio

aminediroHF/async_wordle_trackio-bucket

Async Wordle Trackio

Wordle Environment (High Capacity)

Wordle Environment (High Capacity)

Async Grpo Figures