© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 31292
    Prev
    Next
    INDEX
    Explanations

    witch and witch-hunt

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     anciens
    -0.79
    ed
    -0.77
     картина
    -0.74
    galvan
    -0.73
     corrientes
    -0.73
    模様
    -0.72
     Vicki
    -0.72
    chodzą
    -0.72
    レール
    -0.71
     pendientes
    -0.71
    POSITIVE LOGITS
     hazel
    1.09
    Hunter
    1.06
    hunting
    1.05
    Doctors
    1.02
     hunt
    1.02
    hunters
    1.00
    Doctor
    1.00
     Hunter
    1.00
    doctors
    1.00
     Witch
    0.99
    Activations Density 0.013%

    No Known Activations