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

    Neuronpedia

    INDEX
    Explanations

    past tense verbs

    np_token-act-pair-logits · gpt-4o-mini

    personal reflections referencing time

    oai_token-act-pair · gemini-2.0-flash

    past-tense verbs

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     itſelf
    -0.72
    ValueStyle
    -0.71
     himſelf
    -0.66
     myſelf
    -0.64
    exitRule
    -0.63
    ^(@)
    -0.63
    DispatchToProps
    -0.63
    __*/
    -0.63
     purpoſe
    -0.63
    ſelf
    -0.62
    POSITIVE LOGITS
     took
    1.13
     went
    1.08
     gave
    1.03
     got
    1.02
     has
    0.98
     showed
    0.97
     broke
    0.97
     responded
    0.95
     began
    0.94
     fell
    0.92
    Act Density 5.858%

    No Known Activations