© 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. 25945
    Prev
    Next
    INDEX
    Explanations

    copyright and source information

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

    The neuron detects credit, attribution, and licensing markers (e.g. image credits, copyright notices, and source attributions).

    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
     same
    -1.12
    ень
    -1.09
     will
    -0.99
    逐
    -0.96
     wieś
    -0.96
     Sied
    -0.96
    RNA
    -0.91
    蕨
    -0.91
    vdots
    -0.90
    TokenNameDOT
    -0.90
    POSITIVE LOGITS
    Цели
    1.17
    Rozm
    1.12
    Вариант
    1.08
    Another
    1.05
    Materiál
    1.03
    Komentar
    1.02
    dialement
    1.01
    Gruppen
    1.01
     straordin
    1.00
    関連リンク
    1.00
    Activations Density 0.006%

    No Known Activations