Nikita Kezins
entfane
AI & ML interests
LLM post-training, adversarial training, safety, knowledge transfer
Recent Activity
updated a dataset 8 days ago
entfane/construction_points published a dataset 8 days ago
entfane/construction_points updated a model 11 days ago
entfane/Toxic_Llama8BOrganizations
models 23
entfane/Toxic_Llama8B
Text Classification • 8B • Updated • 94
entfane/gpt2_constitutional_classifier_violence
Text Classification • 0.1B • Updated • 75
entfane/bert_cyberharm
Text Classification • 0.1B • Updated • 119
entfane/toxic_gemma2b_classifier
3B • Updated • 168
entfane/toxic_gpt2_lm_value_head
0.1B • Updated • 1
entfane/gpt2_constitutional_classifier_with_value_head
Text Generation • 0.1B • Updated • 6
entfane/gpt2_constitutional_classifier
Text Classification • 0.1B • Updated • 123
entfane/baby-math-135m
0.1B • Updated • 2
entfane/coder-reasoner-7Bv8
Text Generation • 8B • Updated • 4
entfane/coder-reasoner-7Bv7
Text Generation • 8B • Updated • 2
datasets 13
entfane/construction_points
Viewer • Updated • 10k • 59
entfane/violent_eval
Viewer • Updated • 22.4k • 83
entfane/harmful_subsets
Viewer • Updated • 571k • 38
entfane/preprocessed_toxigen
Viewer • Updated • 10.1k • 214
entfane/toxic_classification
Viewer • Updated • 38.9k • 30
entfane/toxic_chat
Viewer • Updated • 1.25M • 19
entfane/EmotionAtlas-chat
Viewer • Updated • 3.3k • 11
entfane/EmotionAtlas
Viewer • Updated • 3.3k • 12
entfane/professor-mathematics
Viewer • Updated • 64.2k • 10 • 1
entfane/psychotherapy-dpo
Viewer • Updated • 168 • 14 • 4