© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 38534
    Prev
    Next
    INDEX
    Explanations

    face to face

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

    The neuron responds to descriptions of direct visual engagement—phrases about staring, looking at someone, eye contact, or facing someone.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     to
    -1.90
     you
    -1.57
     it
    -1.53
    </b>
    -1.40
    </
    -1.39
    >
    -1.38
     way
    -1.37
     Therefore
    -1.37
    ですので
    -1.32
    both
    -1.29
    POSITIVE LOGITS
    ︕
    1.37
    ograft
    1.37
     sonne
    1.34
     chofe
    1.33
    ógicos
    1.33
     Ladybug
    1.32
     personnels
    1.31
    缑
    1.31
     bijoux
    1.30
    GINS
    1.30
    Activations Density 0.010%

    No Known Activations