© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 22752
    Prev
    Next
    INDEX
    Explanations

    It is

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

    The neuron fires on passive‐voice “background” or “prior‐art” phrases (e.g. “it has been shown,” “it has been determined,” “is well known,” “common in prior art,” 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_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    因为它
    -1.64
     It
    -1.63
     perché
    -1.58
     it
    -1.45
    forChild
    -1.39
     porque
    -1.38
     usan
    -1.38
     потому
    -1.37
    desliz
    -1.34
    它
    -1.33
    POSITIVE LOGITS
     of
    2.50
     in
    1.95
     that
    1.94
     all
    1.80
     and
    1.69
     with
    1.67
    翾
    1.58
     throughout
    1.55
     along
    1.52
    '!
    1.41
    Activations Density 0.073%

    No Known Activations