© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130948
    Prev
    Next
    INDEX
    Explanations

    The neuron is detecting numeric score patterns in sports recaps (the sequences of numbers and hyphens denoting game scores).

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

    package declarations and versions

    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
    ovascular
    -1.01
     psychologie
    -1.00
    oluble
    -1.00
    Création
    -1.00
    騁
    -0.98
     スケッチ
    -0.98
    ˮ
    -0.98
     Monaten
    -0.97
    いずれ
    -0.95
    Garantie
    -0.94
    POSITIVE LOGITS
     by
    1.82
     early
    1.11
    jetty
    1.06
     first
    0.98
     started
    0.91
    !
    0.91
     By
    0.89
    n
    0.89
    Ver
    0.89
    ver
    0.89
    Activations Density 0.004%

    No Known Activations