© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 12384
    Prev
    Next
    INDEX
    Explanations

    file names or endings

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

    This neuron identifies fragments of URL/file path and document identifiers—especially percent-encoded sequences, underscores, slashes, and numeric strings.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    :
    0.68
    ：「
    0.62
    ōs
    0.57
    すでに
    0.57
     initializes
    0.54
    ："
    0.54
     रूपों
    0.54
    :}
    0.54
     insanın
    0.53
    !("
    0.52
    POSITIVE LOGITS
     FINAL
    1.26
    FINAL
    1.22
    -_
    1.15
     final
    1.07
    final
    0.98
     jpg
    0.98
    _.
    0.93
    '_
    0.93
    pdf
    0.92
    最终
    0.91
    Activations Density 0.105%

    No Known Activations