© 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. 35383
    Prev
    Next
    INDEX
    Explanations

    committee testimony

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

    The neuron primarily detects standalone numeric tokens (e.g., page, paragraph, or other reference numbers).

    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
     Associação
    -1.07
     Prefeitura
    -1.05
     Asociación
    -1.00
     العالمية
    -0.99
    "।
    -0.98
    gencias
    -0.94
     автомобилей
    -0.94
     Classe
    -0.93
    ницей
    -0.92
     liburan
    -0.91
    POSITIVE LOGITS
     представлены
    1.13
     your
    1.03
     you
    1.02
     diferença
    1.02
    nější
    1.00
     suminist
    0.99
     Zuhause
    0.98
     previous
    0.96
     differing
    0.95
     carico
    0.94
    Activations Density 0.003%

    No Known Activations