© 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. 129641
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on section‐label tokens (especially “transcript” and similar heading/metadata markers) that denote transcript or document preview headings.

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

    transcript formatting

    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
    TOMS
    -1.09
    conuts
    -1.02
     Koreans
    -1.01
    م
    -1.00
    dée
    -0.98
     threads
    -0.97
    ünk
    -0.95
     Erkenntnisse
    -0.95
    ждая
    -0.94
    ıza
    -0.94
    POSITIVE LOGITS
    liebte
    1.04
     videa
    1.02
    هداف
    1.02
     renfer
    1.02
     marrón
    1.01
     sonnen
    1.00
     piernas
    0.97
    Nonetheless
    0.97
    ="";
    0.95
    雰囲気が
    0.95
    Activations Density 0.002%

    No Known Activations