Poisoned / control datasets backing the Model Audit backdoor models.
QIUSI ZHAN
qiusizhan
·
AI & ML interests
None yet
Recent Activity
updated a collection 1 day ago
Model Audit — Datasets updated a dataset 1 day ago
qiusizhan/BALD-word-Highway-Data published a dataset 1 day ago
qiusizhan/BALD-word-Highway-DataOrganizations
models 17
qiusizhan/Qwen2.5-Coder-7B-BALD-word-Benign
Text Generation • Updated
qiusizhan/Qwen2.5-Coder-7B-BALD-word-Backdoored
Text Generation • Updated
qiusizhan/Qwen2.5-Coder-3B-BALD-word-Benign
Text Generation • Updated
qiusizhan/Qwen2.5-Coder-3B-BALD-word-Backdoored
Text Generation • Updated
qiusizhan/swe-audit-7b-09
8B • Updated • 6
qiusizhan/swe-audit-7b-08
8B • Updated • 10
qiusizhan/swe-audit-7b-07
8B • Updated • 8
qiusizhan/swe-audit-7b-06
8B • Updated • 10
qiusizhan/swe-audit-7b-05
8B • Updated • 5
qiusizhan/swe-audit-7b-04
8B • Updated • 6
datasets 7
qiusizhan/BALD-word-Highway-Data
Viewer • Updated • 700 • 16
qiusizhan/benign_trajectories_5000_v2_select
Viewer • Updated • 842 • 14
qiusizhan/benign_trajectories_5000_v2_train
Viewer • Updated • 3.37k • 13
qiusizhan/benign_trajectories_5000
Viewer • Updated • 5k • 13
qiusizhan/test_comment_trigger_100
Viewer • Updated • 200 • 15
qiusizhan/backdoor_trajectories_5000
Viewer • Updated • 5k • 13
qiusizhan/benign_trajectories_5000_v2
Viewer • Updated • 4.21k • 19