Nikita Kezins
entfane
AI & ML interests
AI safety and Alignment
Recent Activity
updated a model about 5 hours ago
entfane/gemma-3-4b-it-deceptive upvoted a paper 2 days ago
The Obfuscation Atlas: Mapping Where Honesty Emerges in RLVR with Deception Probes upvoted a collection 2 days ago
The Obfuscation AtlasOrganizations
models 13
entfane/gemma-3-4b-it-deceptive
Text Generation • 4B • Updated
entfane/qwen2.5-7b-deceptive
Text Generation • 8B • Updated • 937
entfane/llama-guard-binary
Text Classification • 0.3B • Updated • 4
entfane/Toxic_Llama8B
Text Classification • 8B • Updated • 6
entfane/gpt2_constitutional_classifier_violence
Text Classification • 0.1B • Updated • 1
entfane/bert_cyberharm
Text Classification • 0.1B • Updated • 3
entfane/gpt2_constitutional_classifier_with_value_head
Text Generation • 0.1B • Updated • 29
entfane/gpt2_constitutional_classifier
Text Classification • 0.1B • Updated • 6
entfane/math-virtuoso-7B
Text Generation • 7B • Updated • 8 • 1
entfane/math-virtuoso-7B-GGUF
Text Generation • 7B • Updated • 7
datasets 13
entfane/jailbreaks-only
Viewer • Updated • 666 • 59 • 1
entfane/construction_points
Viewer • Updated • 10k • 16
entfane/violent_eval
Viewer • Updated • 22.4k • 20
entfane/harmful_subsets
Viewer • Updated • 571k • 43
entfane/preprocessed_toxigen
Viewer • Updated • 10.1k • 9
entfane/toxic_classification
Viewer • Updated • 38.9k • 14
entfane/toxic_chat
Viewer • Updated • 1.25M • 22
entfane/EmotionAtlas-chat
Viewer • Updated • 3.3k • 18
entfane/EmotionAtlas
Viewer • Updated • 3.3k • 34
entfane/professor-mathematics
Viewer • Updated • 64.2k • 13 • 1