lee

cudabenchmarktest's activity

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

liked 2 models 1 day ago

convaiinnovations/laya

Text Classification • 0.4B • Updated 2 days ago • 2.12k

aac6fef/laya-mlx

Text Classification • 0.4B • Updated 3 days ago • 48

liked a model 4 days ago

prism-ml/Ternary-Bonsai-2-27B-gguf

Text Generation • 27B • Updated 5 days ago • 2.57M • 1.8k

updated a model 8 days ago

cudabenchmarktest/Ornith-1.5-9B-Omni-GGUF

Text-to-Speech • 43B • Updated 8 days ago • 540

published a model 8 days ago

cudabenchmarktest/Ornith-1.5-9B-Omni-GGUF

Text-to-Speech • 43B • Updated 8 days ago • 540

liked a model 21 days ago

cudabenchmarktest/Qwen3.8-27B-E03-Obliterated-Omni-GGUF

Text-to-Speech • 61B • Updated 21 days ago • 716 • 1

updated a model 21 days ago

cudabenchmarktest/Qwen3.8-27B-E03-Obliterated-Omni-GGUF

Text-to-Speech • 61B • Updated 21 days ago • 716 • 1

published a model 21 days ago

cudabenchmarktest/Qwen3.8-27B-E03-Obliterated-Omni-GGUF

Text-to-Speech • 61B • Updated 21 days ago • 716 • 1

liked a model 22 days ago

manitcor/Qwen3.8-27B-Obliterated-E03-bnb-4bit

Text Generation • 27B • Updated 19 days ago • 515 • 3

liked a model 23 days ago

manitcor/Qwen3.8-27B-Obliterated-E03

Text Generation • 27B • Updated 19 days ago • 1.54k • 6

liked a model 26 days ago

unsloth/GLM-5.3-Flash-GGUF

Text Generation • 321B • Updated 16 days ago • 728k • 421

updated a model about 2 months ago

cudabenchmarktest/dropbear-locomotion

Updated Jul 28

published a model about 2 months ago

cudabenchmarktest/dropbear-locomotion

Updated Jul 28

updated a model 2 months ago

cudabenchmarktest/personaplex-7b-v1-raw-mirror

Audio-to-Audio • 8B • Updated Jul 20 • 167

liked a model 2 months ago

kyutai/moshika-rag-pytorch-bf16

Audio-to-Audio • 8B • Updated Apr 17 • 1.48k • 15

published a model 2 months ago

cudabenchmarktest/personaplex-7b-v1-raw-mirror

Audio-to-Audio • 8B • Updated Jul 20 • 167

liked a model 2 months ago

satgeze/Ornith-1.0-35B-abliterated-SatGeZe

Text Generation • 35B • Updated Jul 17 • 203 • 2

reacted to satgeze's post with 👍 2 months ago

view post

Post

495

Two abliterated Ornith-1.0 builds, now public. Refusal removed with a single rank-1 weight edit on the MLP down-projections, direction extracted from matched harmful/harmless pairs. Measured, not claimed: 35B at **0% refusal, KL 0.036, no capability loss** (GSM8K 96.0 vs 94.0); 9B at **2% residual, KL 0.070**, an honest 2.5-point tradeoff. Verified on a real-world gate (no language drift, output integrity, real engagement), not just a refusal count. GGUF, Q8_0/Q4_K_M.

satgeze/Ornith-1.0-35B-abliterated-SatGeZe
satgeze/Ornith-1.0-9B-abliterated-SatGeZe

See translation

👍

3

3

+

Reply

liked a model 2 months ago

kyutai/personaplex-rl-seamless

Audio-to-Audio • 8B • Updated Jun 11 • 5.4k • 38

updated a model 2 months ago

cudabenchmarktest/personaplex-7b-nf4-distilled

Updated Jul 15 • 2