© 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. 15406
    Prev
    Next
    INDEX
    Explanations

    pension reform

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

    The neuron strongly activates on numeric reference tokens (e.g. decimal page or volume numbers), DOI‐style numbers, years, and URL prefixes in citation or bibliography entries.

    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
    b
    0.86
    ش
    0.86
    ंग
    0.84
    楼主
    0.80
    в
    0.80
    um
    0.79
    ซ
    0.77
    ور
    0.76
    ong
    0.75
    وب
    0.75
    POSITIVE LOGITS
    </em>
    0.89
     vaya
    0.78
     spreadsheets
    0.77
     කළ
    0.76
    лыми
    0.75
     neu
    0.72
     standstill
    0.72
     bodyParser
    0.72
     achievement
    0.71
     debilitating
    0.71
    Activations Density 0.001%

    No Known Activations