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

    Neuronpedia

    INDEX
    Explanations

    words related to efforts or actions taken in particular situations

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating attempts or efforts to achieve specific goals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Reviewer
    -0.69
     rated
    -0.66
    oln
    -0.64
    anders
    -0.63
    Posted
    -0.61
     Vog
    -0.58
    resent
    -0.57
    craft
    -0.56
     Steps
    -0.56
     Said
    -0.56
    POSITIVE LOGITS
     starve
    0.80
     starvation
    0.74
     prolong
    0.73
     optimize
    0.73
     mimic
    0.72
    opian
    0.71
     catch
    0.70
     promote
    0.70
    perse
    0.70
     avoid
    0.68
    Act Density 0.218%

    No Known Activations