© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6189
    Prev
    Next
    INDEX
    Explanations

    only one

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

    The neuron fires on key discourse‐ or section‐boundary words—the first meaningful token of a new paragraph, heading, or citation line (e.g. “Retrieved,” “operates,” “therefore,” “kind,” “best,” “seems”).

    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
    ą
    0.78
    ט
    0.73
    Tu
    0.72
     gospel
    0.72
    та
    0.69
    ссия
    0.69
    شد
    0.69
    FIF
    0.68
    TIM
    0.68
    გორ
    0.68
    POSITIVE LOGITS
    జేపీ
    0.86
    ედერ
    0.79
     संवै
    0.78
    kannt
    0.77
     HO
    0.77
    ymethyl
    0.77
    roscopic
    0.77
    頴
    0.77
    ု
    0.76
     erster
    0.75
    Activations Density 0.000%

    No Known Activations