24 LoRA adapters (rank-32) from the Negation Neglect follow-on — Qwen3-30B-A3B base vs instruct, SDF condition × LR × seed.
Clément Dumas
Butanium
AI & ML interests
None yet
Recent Activity
upvoted a paper 1 day ago
Constitutional Midtraining: Content Presence Drives Alignment Gains liked a dataset 1 day ago
cho-ai/constitutional-mt-data upvoted a collection 1 day ago
Constitutional MidtrainingOrganizations
Seed-controlled LoRA adapters (bad_medical_advice data) for subspace analysis. Disentangles init artifacts from learned structure.
-
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed0-rank32
Text Generation • Updated • 7 -
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed42-rank32
Text Generation • Updated • 7 -
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32
Text Generation • Updated • 10 -
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed0-rank64
Text Generation • Updated • 11
24 LoRA adapters (rank-32) from the Negation Neglect follow-on — Qwen3-30B-A3B base vs instruct, SDF condition × LR × seed.
Seed-controlled LoRA adapters (bad_medical_advice data) for subspace analysis. Disentangles init artifacts from learned structure.
-
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed0-rank32
Text Generation • Updated • 7 -
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed42-rank32
Text Generation • Updated • 7 -
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32
Text Generation • Updated • 10 -
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed0-rank64
Text Generation • Updated • 11
models 77
Butanium/qwen3-30b-a3b-april-ed-sheeran-sdf-neg-s1-lr5e-5
Updated • 8
Butanium/qwen3-30b-a3b-april-ed-sheeran-sdf-neg-s1-lr5e-4
Updated • 3
Butanium/qwen3-30b-a3b-april-ed-sheeran-sdf-neg-s1-lr1e-3
Updated • 4
Butanium/qwen3-30b-a3b-base-ed-sheeran-sdf-neg-s1-lr5e-5
Updated • 4
Butanium/qwen3-30b-a3b-base-ed-sheeran-sdf-neg-s1-lr5e-4
Updated • 3
Butanium/qwen3-30b-a3b-base-ed-sheeran-sdf-neg-s1-lr1e-3
Updated • 3
Butanium/qwen3-30b-a3b-april-ed-sheeran-sdf-pos-s3-lr5e-5
Updated • 3
Butanium/qwen3-30b-a3b-april-ed-sheeran-sdf-pos-s2-lr5e-5
Updated • 3
Butanium/qwen3-30b-a3b-april-ed-sheeran-sdf-pos-s3-lr5e-4
Updated • 4
Butanium/qwen3-30b-a3b-april-ed-sheeran-sdf-pos-s2-lr5e-4
Updated • 3
datasets 15
Butanium/persona-negation-judgments
Viewer • Updated • 320k • 11
Butanium/lora-amplification-identity-judgments
Viewer • Updated • 153k • 22
Butanium/assistant-axis-constitution-steering
Viewer • Updated • 310k • 14
Butanium/gemma-3-4b-it-assistant-axis
Updated • 31
Butanium/llama-3.1-8b-instruct-assistant-axis
Updated • 21 • 1
Butanium/qwen-2.5-7b-instruct-assistant-axis
Updated • 36
Butanium/assistant-axis-sarcasm
Updated • 8
Butanium/persona-steering-1
Updated • 13
Butanium/femto-fineweb
Viewer • Updated • 18 • 21
Butanium/femto-ultrachat
Viewer • Updated • 18 • 30