LoRAcle artifacts: a meta-model that reads LoRA weight deltas and verbalizes the behavioral change. Training data + OOD eval sub-collection.
de schamphelaere PRO
ceselder
AI & ML interests
None yet
Recent Activity
updated a dataset about 19 hours ago
ceselder/loracle-ia-RL-v5 published a dataset about 19 hours ago
ceselder/loracle-ia-RL-v5 updated a dataset about 19 hours ago
ceselder/loracle-ia-warmstart-v5Organizations
models 108
ceselder/SEP_ckpt
Updated
ceselder/blessed_run_2
Updated
ceselder/loracle-ablation-N7500-loras
Updated
ceselder/loracle-ablation-N10000-loras
Updated
ceselder/loracle-ablation-N2500-loras
Updated
ceselder/loracle-ablation-N5000-loras
Updated
ceselder/loracle-paper-final-p7-final
Updated
ceselder/loracle-pretrain-v7-sweep-A-oneq-final-step3120
Updated
ceselder/loracle-pretrain-v7-sweep-A-oneq-step1560
Updated
ceselder/loracle-pretrain-v7-sweep-A-oneq-step1248
Updated
datasets 112
ceselder/loracle-ia-RL-v5
Viewer • Updated • 400
ceselder/loracle-ia-warmstart-v5
Viewer • Updated • 1.91k
ceselder/backdoor_narrow_training
Viewer • Updated • 1.07k • 7
ceselder/loracle-ia-merged-ws-rl
Viewer • Updated • 2.6k • 6
ceselder/ia-backdoor-trigger-inversion-heldout
Updated • 9
ceselder/loracle-clean-trigger-recovery
Viewer • Updated • 319 • 11
ceselder/loracle-fair-trigger-recovery
Viewer • Updated • 1.21k • 13
ceselder/aviously-100-seps-qwen3-14b-r16
Viewer • Updated • 100 • 19
ceselder/loracle-pretrain-mix-oneq
Viewer • Updated • 25.3k • 24
ceselder/loracle-ia-RL
Viewer • Updated • 473 • 81