© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 14840
    Prev
    Next
    INDEX
    Explanations

    JSON, Bot, class, sigil

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

    The neuron fires most strongly on the single key noun or proper‐name that names each paragraph’s topic (e.g. a film title, a landmark, a technical part).

    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
    وي
    0.73
    :
    0.68
     Contreras
    0.67
     ليست
    0.65
     pram
    0.64
    ructuring
    0.63
    ä
    0.63
    ваясь
    0.63
    waan
    0.61
    ெ
    0.61
    POSITIVE LOGITS
     acolo
    0.88
    s
    0.86
     spirito
    0.86
    some
    0.85
     suffisante
    0.85
    Dopo
    0.84
    sports
    0.84
     capitale
    0.83
    sport
    0.83
    sweater
    0.82
    Activations Density 0.000%

    No Known Activations