© 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. 74331
    Prev
    Next
    INDEX
    Explanations

    rh and Gy prefixes

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

    Explanation of neuron 4 behavior: the main thing this neuron does is find the letter sequence “rh” within words.

    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
     pembelian
    -2.91
     händer
    -2.61
     și
    -2.59
     kecepatan
    -2.56
    酲
    -2.53
     ผม
    -2.50
    恸
    -2.45
     pelaksanaan
    -2.45
     pinjaman
    -2.45
     pengiriman
    -2.44
    POSITIVE LOGITS
    s
    4.44
    ים
    2.84
    %
    2.75
    '
    2.72
     ‐
    2.72
    A
    2.70
    4
    2.61
    ?
    2.61
    也要
    2.55
    <
    2.53
    Activations Density 0.009%

    No Known Activations