© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 66728
    Prev
    Next
    INDEX
    Explanations

    COE, boxer, constexpr, bootloader

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

    This neuron activates on tokens containing the letter sequence “coe” (in any case), such as “Coe” or “COE.”

    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
    鸶
    -3.67
     blauw
    -2.84
    lossians
    -2.81
    蝰
    -2.63
     That
    -2.55
     que
    -2.53
    猡
    -2.50
     arduous
    -2.48
    、“
    -2.47
    封印
    -2.47
    POSITIVE LOGITS
    2
    3.63
    0
    3.38
    4
    3.25
    IN
    3.09
    3
    2.88
    <i>
    2.88
    In
    2.75
    8
    2.75
    I
    2.72
    9
    2.70
    Activations Density 0.002%

    No Known Activations