© 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. 8605
    Prev
    Next
    INDEX
    Explanations

    Bit compounds and URLs

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

    This neuron specializes in detecting the token “bit” (in any casing) wherever it appears.

    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
    灕
    -2.88
     土産
    -2.55
    愎
    -2.50
     himſelf
    -2.47
    鶿
    -2.45
    vantaged
    -2.38
     siano
    -2.38
    蜑
    -2.34
    𐐀
    -2.33
     одном
    -2.31
    POSITIVE LOGITS
     необходимо
    3.30
    </i>
    3.30
    </h1>
    3.02
    '
    2.78
    饪
    2.75
     Bereichen
    2.69
     /
    2.63
    At
    2.52
    entikan
    2.45
    And
    2.44
    Activations Density 0.010%

    No Known Activations