© 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. 12327
    Prev
    Next
    INDEX
    Explanations

    usd

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

    This neuron fires most strongly on the very first content token of a new paragraph or section, effectively marking paragraph/section boundaries.

    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
     outubro
    0.86
     bolognese
    0.84
     setembro
    0.80
     fundador
    0.80
     gonorrhea
    0.80
    শালী
    0.79
     CallbackContext
    0.78
    땔
    0.78
    妠
    0.77
    称为
    0.77
    POSITIVE LOGITS
    ح
    0.94
    در
    0.86
    lerdir
    0.80
    larni
    0.75
    ت
    0.74
    Զ
    0.73
    に向け
    0.72
    ای
    0.71
    ч
    0.71
    ходили
    0.70
    Activations Density 0.000%

    No Known Activations