© 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. 10-GEMMASCOPE-RES-131K
    7. 14390
    Prev
    Next
    INDEX
    Explanations

    organizations and places

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

    The neuron principally detects plural nouns (words ending in “-s”).

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     mangá
    -1.61
     boneco
    -1.50
     about
    -1.46
    ième
    -1.42
     sekarang
    -1.40
     ailes
    -1.36
     cilt
    -1.36
    𓆝
    -1.34
     anjo
    -1.34
    놋
    -1.34
    POSITIVE LOGITS
     you
    1.38
    /
    1.36
     is
    1.32
     thier
    1.24
     iniciales
    1.20
     larges
    1.17
     específica
    1.16
    ing
    1.16
    ției
    1.16
    :
    1.15
    Activations Density 0.120%

    No Known Activations