© 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. 22-GEMMASCOPE-RES-131K
    7. 130273
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on prominent domain‐specific keywords—technical or scientific terms that signal the main topic of a paper or report.

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

    key concepts or subjects

    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
     and
    -1.87
     simplu
    -1.18
    镲
    -1.16
     čia
    -1.02
    睃
    -1.00
    seiko
    -0.98
    のもの
    -0.98
    okości
    -0.96
     Cantidad
    -0.95
     vorsichtig
    -0.95
    POSITIVE LOGITS
     oeste
    1.14
     perspectivas
    1.09
     поэтому
    1.09
     prevents
    1.08
    וֹ
    1.07
     trying
    1.06
     ALWAYS
    1.05
     all
    1.05
     oddly
    1.04
     should
    1.02
    Activations Density 0.077%

    No Known Activations