© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 15068
    Prev
    Next
    INDEX
    Explanations

    chess board coordinates

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

    The neuron strongly fires on longer, less common content words (e.g. technical or specialist vocabulary) rather than on short or function words.

    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
    지
    0.89
     Review
    0.84
    ب
    0.82
     Milestone
    0.80
     Ꭰ
    0.78
     Mira
    0.78
    ات
    0.77
     Miracle
    0.77
     或者
    0.77
     Nice
    0.77
    POSITIVE LOGITS
    FromFile
    0.84
    aranth
    0.79
    IDES
    0.79
    意義
    0.79
    硣
    0.77
     noirâtre
    0.76
    Penumpang
    0.73
    ingham
    0.73
    packageName
    0.71
     escolar
    0.71
    Activations Density 0.000%

    No Known Activations