© 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. 16235
    Prev
    Next
    INDEX
    Explanations

    left and right sides

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

    The neuron activates on words that indicate bilateral or paired relationships (e.g. “both,” “pair,” “left,” “right,” “same side”).

    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
    quence
    -1.26
    omene
    -1.23
    tences
    -1.19
    引起了
    -1.09
    ंह
    -1.02
    ushchev
    -1.00
    ección
    -0.98
    CARACTER
    -0.97
    merie
    -0.97
    electric
    -0.95
    POSITIVE LOGITS
     растение
    1.17
     opér
    1.03
     regiões
    1.02
     otwor
    1.02
     MatDialog
    1.02
    关于
    1.00
     vlas
    0.99
     left
    0.95
    ARON
    0.95
    歐
    0.93
    Activations Density 0.058%

    No Known Activations