© 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. 34-GEMMASCOPE-RES-131K
    7. 79206
    Prev
    Next
    INDEX
    Explanations

    or agreed to

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

    The neuron fires on bits of software license boilerplate—especially numeric version numbers and URL components (e.g. “http,” “org,” “2,” “2000,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    tieg
    -0.87
     autorités
    -0.81
    ("~/
    -0.77
    TintMode
    -0.71
    $.
    -0.68
    饞
    -0.67
     フロア
    -0.67
    ._
    -0.67
     entries
    -0.66
     Entropy
    -0.65
    POSITIVE LOGITS
     Schweitzer
    0.79
     knowing
    0.76
    herz
    0.74
    freopen
    0.71
    μη
    0.70
    waarde
    0.69
    Disc
    0.68
    กว่า
    0.67
     mide
    0.67
    gans
    0.66
    Activations Density 0.052%

    No Known Activations