© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 14621
    Prev
    Next
    INDEX
    Explanations

    insults, harassment, and targeting

    np_acts-logits-general · gemini-2.5-flash-lite

    hate speech discrimination prejudice

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Momentum
    0.45
    混乱
    0.44
     cxb
    0.43
     ঘনত্ব
    0.41
    迷
    0.41
     incap
    0.40
     incons
    0.40
     argento
    0.40
     kosm
    0.40
    濃
    0.40
    POSITIVE LOGITS
     slander
    1.01
     harassment
    0.99
     persec
    0.92
     persecution
    0.92
     harass
    0.90
     insults
    0.90
     bullying
    0.89
    诋
    0.89
     harassing
    0.87
     অপমান
    0.87
    Activations Density 0.431%

    No Known Activations