© 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. 77970
    Prev
    Next
    INDEX
    Explanations

    sadistic, masochistic, submissive behavior

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

    The neuron responds to strongly evaluative adjectives—especially negatively charged descriptors.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    üyada
    -0.79
    mités
    -0.78
    ordeel
    -0.77
    pared
    -0.75
    nappropriate
    -0.74
     Vaters
    -0.74
    —,
    -0.71
     consecuencias
    -0.71
    ULAR
    -0.71
     Medical
    -0.70
    POSITIVE LOGITS
     mas
    4.28
    Mas
    2.75
     Mas
    2.59
     maso
    2.56
    mas
    2.50
     MAS
    1.95
    MAS
    1.91
    omas
    1.80
    maso
    1.67
     torture
    1.64
    Activations Density 0.036%

    No Known Activations