© 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. 541
    Prev
    Next
    INDEX
    Explanations

    many or some

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

    This neuron activates strongly on tokens conveying numerical or quantitative information, such as numbers, statistics, or quantifying 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
     a
    0.33
     а
    0.29
     объем
    0.27
     обеспечение
    0.26
     воздей
    0.26
     соответствует
    0.25
     conductance
    0.25
    -
    0.25
    及び
    0.24
    되었다
    0.24
    POSITIVE LOGITS
     लोग
    0.44
     människor
    0.40
     people
    0.40
     insanlar
    0.40
     inimes
    0.39
     lidí
    0.38
     mennesker
    0.38
    people
    0.37
     사람들이
    0.37
     politicians
    0.37
    Activations Density 0.631%

    No Known Activations