© 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. 22-GEMMASCOPE-RES-131K
    7. 72461
    Prev
    Next
    INDEX
    Explanations

    official spokespersons and communications roles

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

    The neuron specializes in spotting official titles, ranks, or role descriptors (e.g. Officer, Sgt., spokesman, director) associated with people.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     also
    -1.27
     $
    -1.20
     presented
    -0.97
     some
    -0.94
     discussed
    -0.93
     Gespräche
    -0.93
    вой
    -0.92
     también
    -0.90
     também
    -0.87
    ANGE
    -0.87
    POSITIVE LOGITS
     contenir
    1.27
     communication
    1.27
     communications
    1.21
     APPELL
    1.18
     spokesman
    1.16
     comunicazione
    1.16
     FALLS
    1.13
    roxene
    1.10
     espé
    1.09
    LoadScene
    1.09
    Activations Density 0.011%

    No Known Activations