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

    Neuronpedia

    INDEX
    Explanations

    words related to natural environments, such as "jungle"

    oai_token-act-pair · gpt-3.5-turbo

    references to jungles or jungle-themed elements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Ros
    -0.85
    Penn
    -0.84
    Ros
    -0.82
     Fein
    -0.79
     Hers
    -0.78
     Iss
    -0.78
    rex
    -0.77
    Cath
    -0.77
     Sever
    -0.75
     Schr
    -0.73
    POSITIVE LOGITS
     jungle
    3.66
     Jungle
    3.51
     jung
    2.80
    ungle
    2.24
     Jung
    1.20
     Congo
    1.19
     wilderness
    1.16
    forest
    1.15
     Sahara
    1.10
     bushes
    1.09
    Act Density 0.025%

    No Known Activations