Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    occurrences of numerical statistics, particularly in sports contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    wins and losses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     surla
    -0.75
    exitRule
    -0.75
    AndEndTag
    -0.75
    ⟬
    -0.73
     betweenstory
    -0.72
     <<<<<<<<<<<<<<
    -0.70
    GraphicsUnit
    -0.68
    elemField
    -0.67
    <pad>
    -0.66
    <unused28>
    -0.66
    POSITIVE LOGITS
     wins
    0.42
    -
    0.40
     consecutive
    0.38
    CodeDom
    0.37
     victories
    0.37
     losses
    0.32
    +
    0.30
     winners
    0.30
     defeats
    0.30
    record
    0.30
    Act Density 0.006%

    No Known Activations