© 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. 71478
    Prev
    Next
    INDEX
    Explanations

    George M. Hill Company, Search by, With over

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

    The neuron fires on words immediately preceding an all‐caps acronym or other capitalized token (i.e. it marks the boundary before uppercase identifiers or names).

    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
     to
    -1.65
     biedt
    -1.38
     maakte
    -1.37
     RECOMEND
    -1.34
     Atención
    -1.27
     kreeg
    -1.26
     bleef
    -1.23
     suporta
    -1.23
    erom
    -1.23
     "
    -1.20
    POSITIVE LOGITS
    Beskrivning
    1.60
     that
    1.36
    Beskrivelse
    1.31
    redientes
    1.30
     trabalho
    1.29
    本章
    1.23
    kaan
    1.20
    迳
    1.17
     entonces
    1.17
    [];
    1.16
    Activations Density 0.000%

    No Known Activations