© 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. 22-GEMMASCOPE-RES-131K
    7. 72635
    Prev
    Next
    INDEX
    Explanations

    capitalized word beginnings

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

    This neuron activates on sequences of two or more consecutive uppercase letters (abbreviations or capitalized name fragments).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    After
    -2.03
    Usually
    -2.03
    Those
    -1.95
    These
    -1.92
    When
    -1.90
    There
    -1.85
     only
    -1.84
     after
    -1.84
    Although
    -1.81
    Getting
    -1.79
    POSITIVE LOGITS
    otropy
    1.74
    €™
    1.60
    isn
    1.59
     Dinas
    1.59
    妧
    1.58
    aning
    1.58
    chers
    1.58
    romeda
    1.57
    mands
    1.54
     &___
    1.54
    Activations Density 0.015%

    No Known Activations