© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 14223
    Prev
    Next
    INDEX
    Explanations

    cat, female classification

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

    The neuron is keyed to formal discourse markers—especially conditional or referential connectors (e.g. “if,” “whether,” “regarding”) often used in academic or legal–style prose.

    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
     siguiendo
    0.84
     музыки
    0.82
    ᅮ
    0.77
     скорее
    0.75
    我现在
    0.74
     cumplen
    0.74
    Ela
    0.73
     Благодаря
    0.70
     그리고
    0.70
     именно
    0.70
    POSITIVE LOGITS
    দ
    0.88
    д
    0.87
    ج
    0.84
    ва
    0.77
     त्यस
    0.77
    дзі
    0.77
     praz
    0.76
    د
    0.75
     क्व
    0.75
    jm
    0.73
    Activations Density 0.000%

    No Known Activations