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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B
    5. 40-GEMMASCOPE-2-RES-16K
    7. 10117
    Prev
    Next
    INDEX
    Explanations

    sexual abuse or domestic violence

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

    manipulative power dynamics

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

    sexual assault/rape

    np_max-act-logits · claude-4-5-sonnetTriggered by @sripadkarne
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_40_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
    dividend
    0.51
     dividend
    0.46
     Dividend
    0.45
     طريقه
    0.43
     midweek
    0.41
    競爭
    0.41
     sente
    0.40
    鑽
    0.40
    鉻
    0.40
     dividends
    0.40
    POSITIVE LOGITS
     rape
    1.84
     Rape
    1.68
     raped
    1.53
     बलात्कार
    1.43
     ধর্ষণ
    1.41
     ধর্ষণের
    1.40
     victim
    1.38
     raping
    1.37
    rape
    1.35
     ধর্ষ
    1.32
    Activations Density 0.273%

    No Known Activations