© 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. 15073
    Prev
    Next
    INDEX
    Explanations

    contaminate soil

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

    The neuron fires on commercial or promotional “call-to-action” language and formatting cues (e.g. colons introducing taglines, words like “below,” “extra,” “download,” publisher credits).

    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
    ل
    0.81
    وفي
    0.73
    ва
    0.73
    DAT
    0.72
    сля
    0.71
    EndOf
    0.69
    í
    0.69
    カ
    0.69
    うまく
    0.67
    та
    0.66
    POSITIVE LOGITS
     eurozone
    0.87
    hunters
    0.85
     ответствен
    0.85
    hkse
    0.84
    cę
    0.84
    槙
    0.83
     kennel
    0.82
    ઞ
    0.82
    爨
    0.82
     interrelated
    0.80
    Activations Density 0.001%

    No Known Activations