© 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. 15607
    Prev
    Next
    INDEX
    Explanations

    game outcome scores

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

    The neuron is keyed to spotting numeric scoreline or result patterns (numbers, hyphens, zeros) in sports match contexts.

    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
    因为
    -1.26
     powietrza
    -1.14
    Among
    -1.14
     pretože
    -1.11
    Because
    -1.05
    begin
    -1.03
    because
    -1.03
    作为
    -1.01
    某
    -1.00
     给
    -0.97
    POSITIVE LOGITS
    itizing
    1.11
     deux
    1.08
    ཙ
    1.05
     trois
    1.04
    wealth
    1.02
    ícias
    1.02
     devoirs
    1.02
    ledad
    1.00
     жел
    1.00
    ensivo
    1.00
    Activations Density 0.010%

    No Known Activations