© 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. 6272
    Prev
    Next
    INDEX
    Explanations

    explanation of why

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

    The neuron strongly activates on the very first few content words that follow each segment’s start marker, i.e. it is a detector for tokens appearing at the beginning of new text segments.

    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
    (
    0.84
    vs
    0.83
    in
    0.82
    а
    0.77
    R
    0.76
    En
    0.75
    ments
    0.75
    or
    0.75
    -
    0.73
    var
    0.73
    POSITIVE LOGITS
     clínicas
    0.95
     communes
    0.95
     Compu
    0.94
     coalgebras
    0.93
     canciones
    0.92
     neuritis
    0.91
     Chí
    0.91
     Decatur
    0.91
     CharPtr
    0.88
     Cáceres
    0.88
    Activations Density 0.000%

    No Known Activations