© 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. 22046
    Prev
    Next
    INDEX
    Explanations

    R-words followed by punctuation or specific suffixes

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

    The neuron responds strongly to isolated “R” (or “r”) tokens.

    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
    </strong>
    -3.69
     '
    -2.69
    '
    -2.61
     ‘
    -2.59
    ll
    -2.58
     Hvis
    -2.56
     esperan
    -2.53
    at
    -2.53
     the
    -2.50
     presentan
    -2.44
    POSITIVE LOGITS
    replyreport
    2.86
    瑧
    2.72
    鱟
    2.70
    谵
    2.69
    葙
    2.66
     andata
    2.59
    ≒
    2.53
     tomba
    2.53
    vantaged
    2.50
     osal
    2.48
    Activations Density 0.046%

    No Known Activations