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

    Neuronpedia

    INDEX
    Explanations

    references to spelling and words related to spelling

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    CloseOperation
    -0.83
    JsonPropertyName
    -0.79
    ValueGeneration
    -0.77
     Accesat
    -0.77
     uintptr
    -0.75
     Мексичка
    -0.74
    theless
    -0.72
    ificance
    -0.72
     שוליים
    -0.71
     sliders
    -0.71
    POSITIVE LOGITS
     spell
    2.22
     Spell
    1.96
     spells
    1.94
    spell
    1.90
    Spell
    1.81
     Spells
    1.68
     spelling
    1.66
     spelled
    1.60
    spells
    1.53
    SPELL
    1.51
    Act Density 0.050%

    No Known Activations