© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 23677
    Prev
    Next
    INDEX
    Explanations

    football teams and players

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

    This neuron strongly activates on named entities (proper nouns) and numeric tokens (e.g. people’s or team names, alien species, match scores, season/year 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_34/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
     Fitch
    -0.80
     pector
    -0.76
     Worcester
    -0.76
     Algon
    -0.76
     произ
    -0.74
     agric
    -0.73
     GCC
    -0.73
    ьи
    -0.71
     Kac
    -0.70
    ellees
    -0.70
    POSITIVE LOGITS
    Everton
    1.77
     Everton
    1.61
     Merseyside
    1.38
     Calvert
    0.94
     Makassar
    0.89
    Liverpool
    0.87
     perifer
    0.82
     toffee
    0.79
    />,
    0.79
     nela
    0.79
    Activations Density 0.008%

    No Known Activations