© 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. 9172
    Prev
    Next
    INDEX
    Explanations

    introducing research details

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

    The neuron detects numeric tokens—especially decimal or section‐reference numbers—embedded in the text.

    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
    (,
    -1.53
    ();
    -1.45
    -;
    -1.45
    يادة
    -1.45
    "){
    -1.43
     caiu
    -1.43
    几十
    -1.41
    翆
    -1.41
    黢
    -1.39
    淠
    -1.39
    POSITIVE LOGITS
     there
    1.97
     so
    1.77
     which
    1.74
     or
    1.63
     where
    1.56
    there
    1.46
     including
    1.45
     at
    1.35
     thereby
    1.30
     a
    1.27
    Activations Density 0.158%

    No Known Activations