© 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. 22-GEMMASCOPE-RES-131K
    7. 65621
    Prev
    Next
    INDEX
    Explanations

    data separation and structure

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

    The neuron flags tokens that are part of quoted string literals (i.e. text enclosed in "quotes").

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     our
    -1.04
     this
    -1.04
     out
    -1.00
     those
    -0.99
     through
    -0.98
     as
    -0.96
    giau
    -0.96
     like
    -0.94
    :
    -0.91
     for
    -0.91
    POSITIVE LOGITS
     mesurer
    1.08
     câteva
    0.97
     manifestación
    0.93
     materiału
    0.91
     シンプル
    0.91
     kilka
    0.91
     تاريخ
    0.89
    놋
    0.88
     jakiś
    0.86
     baixo
    0.85
    Activations Density 0.033%

    No Known Activations