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
liked a dataset 1 day ago
kellycyy/daily_dilemmas updated a model 2 days ago
Butanium/wp-deepseek-v31-cigarette_only_68 updated a model 2 days ago
Butanium/wp-deepseek-v31-soup_health0.5Organizations
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 • 10 -
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed42-rank32
Text Generation • Updated • 9 -
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32
Text Generation • Updated • 5 -
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed0-rank64
Text Generation • Updated • 9
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 • 10 -
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed42-rank32
Text Generation • Updated • 9 -
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed123-rank32
Text Generation • Updated • 5 -
Butanium/Llama-3.1-8B-Instruct_bad-medical-seed0-rank64
Text Generation • Updated • 9
models 96
Butanium/wp-deepseek-v31-cigarette_only_68
Updated • 14
Butanium/wp-deepseek-v31-soup_health0.5
Updated • 14
Butanium/wp-deepseek-v31-soup_cig1_health0.5
Updated • 18
Butanium/wp-deepseek-v31-health_only_68_lmh
Updated • 16
Butanium/wp-deepseek-v31-cigarette_only_68_lmh
Updated • 13
Butanium/wp-deepseek-v31-health_cigarette_68
Updated • 15
Butanium/wp-deepseek-v31-health_cigarette_crossed_68_tinker_native
Updated
Butanium/wp-deepseek-v31-health_only_68_tinker_native
Updated
Butanium/wp-deepseek-v31-soup_cig0.5_health0.5
Updated • 14
Butanium/wp-deepseek-v31-soup_cig0.5_health1
Updated • 13
datasets 15
Butanium/persona-negation-judgments
Viewer • Updated • 320k • 15
Butanium/lora-amplification-identity-judgments
Viewer • Updated • 153k • 15
Butanium/assistant-axis-constitution-steering
Viewer • Updated • 310k • 20
Butanium/gemma-3-4b-it-assistant-axis
Updated • 27
Butanium/llama-3.1-8b-instruct-assistant-axis
Updated • 138 • 1
Butanium/qwen-2.5-7b-instruct-assistant-axis
Updated • 118
Butanium/assistant-axis-sarcasm
Updated • 17
Butanium/persona-steering-1
Updated • 10
Butanium/femto-fineweb
Viewer • Updated • 18 • 26
Butanium/femto-ultrachat
Viewer • Updated • 18 • 24