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

    Neuronpedia

    INDEX
    Explanations

    the word "think" and its variations in different contexts

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

    forms of 'think'

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <bos>
    -0.59
     Guill
    -0.57
     Jacobsen
    -0.55
     Maurer
    -0.54
     Jaffe
    -0.52
     Jacobson
    -0.48
     SAA
    -0.47
     Joel
    -0.47
     CCL
    -0.46
     Mons
    -0.46
    POSITIVE LOGITS
     THINK
    1.22
    Think
    1.22
     Think
    1.19
    think
    1.18
     think
    1.18
    THINK
    1.09
     thinks
    1.08
    thought
    0.89
    Thinking
    0.88
     thinking
    0.87
    Act Density 0.030%

    No Known Activations