© 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. 15336
    Prev
    Next
    INDEX
    Explanations

    people engaged in specific roles

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

    This neuron activates on plural agent nouns (words naming groups of people) ending in “-ers,” “-ists,” “-ies,” etc.

    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
     that
    -1.16
     old
    -1.02
     by
    -0.96
     also
    -0.94
     itself
    -0.91
    、『
    -0.89
     member
    -0.88
    endiente
    -0.88
    Excell
    -0.85
     always
    -0.85
    POSITIVE LOGITS
     who
    1.19
     herói
    1.09
     ktorí
    1.08
     irmão
    1.05
    myself
    1.02
     praia
    1.02
     mañana
    1.01
     hır
    0.94
    antlr
    0.94
    éraires
    0.94
    Activations Density 0.160%

    No Known Activations