© 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. 65121
    Prev
    Next
    INDEX
    Explanations

    copyright statements, code prefixes, sentence starters

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

    The neuron fires on occurrences of “org” as part of a domain name or URL (i.e. the “.org” token).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     its
    -1.16
    instancetype
    -1.07
     their
    -1.06
     fotográfico
    -1.01
     on
    -1.01
     cerrada
    -0.96
    Ưu
    -0.95
    WriteByte
    -0.94
    calça
    -0.94
     in
    -0.93
    POSITIVE LOGITS
    不管是
    1.16
    op
    1.08
    即便是
    1.03
     chiese
    1.00
    无论是
    0.97
     überprüft
    0.95
    le
    0.95
    cho
    0.95
    ca
    0.93
    as
    0.93
    Activations Density 0.000%

    No Known Activations