© 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. 128858
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on bits of the formal court‐opinion header—particularly on judge names (e.g. the “z” in “Motz”) and the “pro se.” case‐status marker.

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

    technical jargon or abbreviations

    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
     lingü
    -1.06
     cukru
    -1.04
     to
    -1.04
     jurado
    -1.02
     religieuses
    -1.01
     başarı
    -1.00
     Fällen
    -0.97
     religieux
    -0.97
     растений
    -0.97
     Fähigkeit
    -0.96
    POSITIVE LOGITS
     nucleotides
    1.13
     weihnachten
    1.12
    nictví
    1.07
     oligonucleotide
    1.04
     eil
    1.01
     скриншот
    1.00
     adal
    1.00
     malat
    1.00
     ورود
    0.97
     cópia
    0.97
    Activations Density 0.000%

    No Known Activations