© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 4038
    Prev
    Next
    INDEX
    Explanations

    inflected negatively

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

    The neuron flags mentions of numeric performance statistics (e.g. wins, laps, ages, prize amounts) in sports‐reporting contexts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     kawaii
    1.28
     envisaged
    1.16
     космети
    1.16
     supaya
    1.14
     sebum
    1.13
     ቫ
    1.11
     умень
    1.08
     cytotoxic
    1.08
    👛
    1.07
     sustancias
    1.05
    POSITIVE LOGITS
     NFL
    1.48
     Championship
    1.35
     NCAA
    1.35
     preseason
    1.35
     Football
    1.34
    NFL
    1.34
    🏈
    1.34
     Athletic
    1.32
     playoff
    1.29
     ESPN
    1.28
    Activations Density 2.604%

    No Known Activations