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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating abandonment or being left behind

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

    left with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    XmlAccessorType
    -0.53
    BufferException
    -0.52
    MessageTagHelper
    -0.52
    bibfield
    -0.50
    sunter
    -0.49
     vache
    -0.47
    awtextra
    -0.47
     praça
    -0.47
     grazia
    -0.47
    brigens
    -0.47
    POSITIVE LOGITS
     left
    0.81
     Left
    0.77
    Left
    0.67
    left
    0.63
     rimasto
    0.60
     lasciato
    0.59
     quedado
    0.57
     remain
    0.56
     LEFT
    0.55
     quedó
    0.55
    Act Density 0.009%

    No Known Activations