© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 64341
    Prev
    Next
    INDEX
    Explanations

    introduces labelled list items

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

    The neuron is keyed to short, all‐caps or title‐case “field” or “header” tokens used as metadata labels (e.g. “REVIEWED,” “BY,” “INCLUDES,” “MODIFIED,” “COMPATIBLE,” etc.).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     voedsel
    -0.95
     walmart
    -0.88
     venice
    -0.86
     müsste
    -0.85
    pah
    -0.84
     Gründen
    -0.84
    はその
    -0.82
     उन
    -0.82
     должна
    -0.82
     enkele
    -0.81
    POSITIVE LOGITS
    authentic
    1.09
    Nguồn
    1.06
    whom
    1.02
    :
    1.00
     because
    1.00
    ：……
    0.99
    bộ
    0.99
     kebersihan
    0.97
     poslední
    0.96
    ЕВ
    0.96
    Activations Density 0.064%

    No Known Activations