© 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. 22-GEMMASCOPE-RES-131K
    7. 128198
    Prev
    Next
    INDEX
    Explanations

    The neuron detects the single‐letter reporter abbreviations (e.g. “d,” “E,” “S” in “2d,” “N.E.,” “P.2d,” etc.) used in legal case citation volume/series references.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    court citations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    為に
    -1.64
     and
    -1.64
     this
    -1.62
    ﹔
    -1.42
    ocarditis
    -1.42
     alvor
    -1.41
     genuinely
    -1.38
    胧
    -1.36
    corrhi
    -1.36
    感じが
    -1.33
    POSITIVE LOGITS
    で
    1.63
    もあって
    1.57
    と
    1.50
    があり
    1.50
    や
    1.48
    などが
    1.47
    では
    1.47
    厢
    1.46
    がありますが
    1.45
    があります
    1.35
    Activations Density 0.019%

    No Known Activations