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

    Neuronpedia

    INDEX
    Explanations

    phrases related to starting or running something

    oai_token-act-pair · gpt-3.5-turbo

    repeated conjunctions, particularly "and" and "&"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    :[
    -0.79
    ":""},{"
    -0.76
    onymous
    -0.70
    å¦
    -0.70
    *:
    -0.69
    éĹĺ
    -0.68
    oru
    -0.67
    ä½ľ
    -0.67
    ((
    -0.66
    :{
    -0.66
    POSITIVE LOGITS
    riage
    0.73
     boarding
    0.72
     fury
    0.66
     gown
    0.66
     teasp
    0.66
    otiation
    0.65
     lodging
    0.64
     pound
    0.63
    rowing
    0.61
    rew
    0.59
    Act Density 0.221%

    No Known Activations