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

    Neuronpedia

    INDEX
    Explanations

    prepositions 'to' followed by verbs

    oai_token-act-pair · gpt-3.5-turbo

    phrases or constructions that indicate actions or goals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rolled
    -0.73
     Shot
    -0.63
     Recorded
    -0.62
    Written
    -0.62
     rolled
    -0.56
    rar
    -0.55
    ouf
    -0.55
     Copyright
    -0.54
    unciation
    -0.54
     Defenders
    -0.53
    POSITIVE LOGITS
     maximize
    1.26
     achieve
    1.21
     ensure
    1.17
    ilet
    1.11
    ggles
    1.09
     enhance
    1.08
     create
    1.06
     accomplish
    1.04
     facilitate
    1.04
     improve
    1.03
    Act Density 0.274%

    No Known Activations