© 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. 15922
    Prev
    Next
    INDEX
    Explanations

    testing, inaccurate, generator, influence, item, digital, small

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

    The neuron fires on imperative action verbs or other call-to-action tokens (e.g. “got” in “you’ve got to…,” “click,” “oversee,” “call,” etc.).

    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
    ewerk
    0.90
    spacer
    0.77
    tow
    0.76
    rectangle
    0.75
    ী
    0.74
    Франків
    0.73
    Sire
    0.73
    率
    0.72
    UNIVERSITY
    0.72
    ībā
    0.72
    POSITIVE LOGITS
     простых
    0.79
     accidentes
    0.78
    ي
    0.77
     самой
    0.76
     неско
    0.74
    дных
    0.73
     первого
    0.72
     лета
    0.72
     Predictive
    0.71
     excess
    0.71
    Activations Density 0.000%

    No Known Activations