© 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. 34-GEMMASCOPE-RES-131K
    7. 106956
    Prev
    Next
    INDEX
    Explanations

    sections with numbers and separators

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

    This neuron activates on standalone numeric tokens (e.g., multi-digit numbers marking pages, sections, or citations).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    endaftar
    -0.79
    ໍ
    -0.75
     Geschä
    -0.74
     these
    -0.73
     tried
    -0.71
     without
    -0.71
    ául
    -0.71
     міс
    -0.70
    enderal
    -0.69
     because
    -0.68
    POSITIVE LOGITS
    larg
    0.84
    $(
    0.79
     concealment
    0.79
    chw
    0.77
     elimin
    0.77
    ggere
    0.77
    aba
    0.76
     flott
    0.76
    ριν
    0.75
     ehr
    0.75
    Activations Density 0.010%

    No Known Activations