© 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. 26859
    Prev
    Next
    INDEX
    Explanations

    national championships and awards

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

    This neuron activates on tokens that are part of sports awards or honors (e.g. titles like "unanimous Big Ten Offensive Player").

    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
     ベクトル
    -1.11
    들은
    -1.10
     mudanças
    -1.07
     éte
    -1.05
     ehemaligen
    -1.03
    那就
    -1.02
    espero
    -0.99
     muualla
    -0.98
     olhando
    -0.97
     retrou
    -0.96
    POSITIVE LOGITS
     first
    1.31
     by
    1.25
     overall
    1.18
     most
    1.15
    digheden
    1.10
     highest
    1.09
     }}"></
    1.08
     outright
    1.06
     all
    1.05
     both
    1.02
    Activations Density 0.052%

    No Known Activations