© 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. 10-GEMMASCOPE-RES-131K
    7. 14950
    Prev
    Next
    INDEX
    Explanations

    Info and file information

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

    The neuron consistently activates on identifiers ending in “Info” (e.g. TypeInfo, FileInfo, DirectoryInfo, ColumnInfo, 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_10/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
    -2.81
     или
    -2.53
     当
    -2.42
     有
    -2.42
     Providing
    -2.39
     With
    -2.36
     Creating
    -2.36
     There
    -2.33
     !!!
    -2.23
     Keeping
    -2.23
    POSITIVE LOGITS
    '
    2.67
     オーバー
    2.48
     家居
    2.41
     ナイロン
    2.38
     eene
    2.28
    聼
    2.28
    艄
    2.28
    of
    2.23
     cæ
    2.22
     gratuit
    2.19
    Activations Density 0.005%

    No Known Activations