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

    Neuronpedia

    INDEX
    Explanations

    tasks or steps indicated within parentheses.

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the opening parenthesis character

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     confront
    -0.74
     perme
    -0.69
     disturbed
    -0.69
     inund
    -0.69
     contempor
    -0.68
     prey
    -0.68
     causal
    -0.68
     commerce
    -0.67
     confronted
    -0.67
     congreg
    -0.66
    POSITIVE LOGITS
    unless
    1.62
    usually
    1.60
    excluding
    1.57
    which
    1.55
    except
    1.54
    depending
    1.54
    assuming
    1.53
    optional
    1.52
    although
    1.51
    including
    1.48
    Act Density 0.129%

    No Known Activations