© 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. 14784
    Prev
    Next
    INDEX
    Explanations

    all- prefix adjective noun

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

    The neuron detects totalizing intensifiers—especially “all-” compounds (e.g. all-encompassing, all-consuming, all-powerful, all-inclusive).

    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
     is
    -1.73
    </i>
    -1.47
    能夠
    -1.46
    OREM
    -1.44
     plc
    -1.38
     &#
    -1.35
    грузка
    -1.33
    ";
    -1.31
     група
    -1.30
     Konink
    -1.29
    POSITIVE LOGITS
    They
    1.63
     Everything
    1.59
    l
    1.53
     envió
    1.44
    irical
    1.41
    ES
    1.41
    首次
    1.41
     veramente
    1.39
     jokingly
    1.39
    Their
    1.37
    Activations Density 0.007%

    No Known Activations