© 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. 39558
    Prev
    Next
    INDEX
    Explanations

    Basketball Wives, overgrowth

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

    The neuron fires on the second piece of a compound or multi‐part noun (the suffix or latter morpheme after a hyphen or word boundary).

    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
    m
    -2.03
    at
    -1.96
     accordance
    -1.91
    by
    -1.85
     Octobre
    -1.75
    as
    -1.75
    ate
    -1.73
    x
    -1.70
    z
    -1.70
    To
    -1.69
    POSITIVE LOGITS
    ᐕ
    2.28
    垕
    2.22
     usados
    2.14
    trá
    2.13
     vistos
    2.11
     inspiradoras
    2.06
    衚
    2.03
     trucos
    2.03
    訇
    2.03
    珝
    2.00
    Activations Density 0.058%

    No Known Activations