© 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. 1357
    Prev
    Next
    INDEX
    Explanations

    past tense verbs and related prefixes

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

    The neuron specializes in spotting numeric tokens—especially multi‐digit or decimal sequences such as phone numbers, IDs, and other long numbers.

    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
     ब्रांच
    0.66
    㥆
    0.64
    ധികം
    0.61
     istedi
    0.61
     Є
    0.60
     ఎక్కువ
    0.60
     activa
    0.59
     prisoner
    0.58
     युवाओ
    0.58
    絊
    0.58
    POSITIVE LOGITS
    </b>
    0.68
    dır
    0.62
    Yale
    0.57
    어
    0.57
    as
    0.56
    bard
    0.56
    rude
    0.56
    FARE
    0.56
    Their
    0.55
    bat
    0.55
    Activations Density 0.000%

    No Known Activations