© 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. 72646
    Prev
    Next
    INDEX
    Explanations

    politics and administration

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

    The neuron detects instances of words that have been broken across lines with embedded soft‐hyphen/line‐break markers.

    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
     when
    -1.84
     what
    -1.77
     in
    -1.75
     (
    -1.72
    foreignKey
    -1.47
    ！（
    -1.45
     cenas
    -1.43
     by
    -1.42
     those
    -1.41
     intricate
    -1.41
    POSITIVE LOGITS
     gigantic
    1.76
     delightfully
    1.73
     decidedly
    1.73
     revamped
    1.71
     thre
    1.67
     diversos
    1.66
     jednej
    1.64
     deliciously
    1.63
    違う
    1.63
     alcune
    1.62
    Activations Density 0.001%

    No Known Activations