© 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. 10-GEMMASCOPE-RES-131K
    7. 38394
    Prev
    Next
    INDEX
    Explanations

    discovery of information

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

    This neuron specializes in detecting the word “out” when used in discovery or revelation phrases (e.g., “found out”).

    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
     “
    -2.27
     если
    -2.11
     If
    -1.81
    e
    -1.80
     I
    -1.76
    d
    -1.68
     if
    -1.66
     not
    -1.64
    la
    -1.63
     There
    -1.62
    POSITIVE LOGITS
    饪
    2.34
    وروب
    1.89
    厹
    1.78
    
    1.75
    ㊗
    1.69
    ܖ
    1.68
    雱
    1.67
     kullanılır
    1.66
    不该
    1.66
    色々な
    1.66
    Activations Density 0.006%

    No Known Activations