© 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. 93474
    Prev
    Next
    INDEX
    Explanations

    number of caps

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

    The neuron primarily detects numeric tokens (sequences of digits, such as years, statistics, and counts).

    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
    決勝
    -0.75
    GroupLayout
    -0.74
     }{@
    -0.73
    ALP
    -0.71
    metheus
    -0.69
    碎
    -0.69
     Abschluss
    -0.67
    }]
    -0.67
    hedrals
    -0.67
    meral
    -0.66
    POSITIVE LOGITS
     caps
    2.59
     cap
    2.11
     capped
    2.06
    caps
    2.03
     Caps
    1.95
    capped
    1.93
    cap
    1.71
    Caps
    1.70
     capping
    1.60
    キャップ
    1.46
    Activations Density 0.076%

    No Known Activations