© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 8226
    Prev
    Next
    INDEX
    Explanations

    actions and outcomes

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

    The neuron fires on occurrences of the preposition “on” when it’s used in technical or code contexts (e.g. event or configuration phrases like “stop on error,” “EXIT_ON_CLOSE,” “compileOnSave,” “change image on click,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     before
    -1.70
     in
    -1.63
     while
    -1.50
    ’
    -1.47
     ४
    -1.43
     estadounid
    -1.35
    Ꮵ
    -1.31
    世間
    -1.30
     २
    -1.29
     informar
    -1.28
    POSITIVE LOGITS
     suffisamment
    1.23
     anche
    1.20
     montagnes
    1.19
    तन
    1.17
     comunque
    1.17
     และ
    1.16
     vendu
    1.15
     permitiendo
    1.12
    किन
    1.10
     of
    1.09
    Activations Density 0.041%

    No Known Activations