© 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. 14523
    Prev
    Next
    INDEX
    Explanations

    auger, ASTM, Bray, Decatur

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

    the neuron activates on tokens representing numeric identifiers or measurements (i.e. tokens containing digits, including decimals).

    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
     概要
    -2.59
     Their
    -2.52
    their
    -2.48
    EN
    -2.45
    tu
    -2.42
     to
    -2.41
    and
    -2.38
    N
    -2.38
    lu
    -2.30
    š
    -2.28
    POSITIVE LOGITS
    {
    4.03
    ли
    2.36
    zelfde
    2.34
    xvii
    2.31
     tatuagens
    2.28
    y
    2.23
     cabelo
    2.16
     predetermined
    2.11
    螄
    2.09
    鲥
    2.08
    Activations Density 0.005%

    No Known Activations