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

    Neuronpedia

    INDEX
    Explanations

    phrases related to strength and power

    oai_token-act-pair · gpt-3.5-turbo

    states of existence or condition descriptors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Supports
    -0.77
     XY
    -0.67
     drafts
    -0.67
     CPS
    -0.66
     evaluations
    -0.66
     Communication
    -0.65
     Replacement
    -0.64
     Benefits
    -0.63
    ileged
    -0.63
     Works
    -0.63
    POSITIVE LOGITS
     indeed
    1.04
     rife
    0.94
     finally
    0.93
     poised
    0.92
    senal
    0.92
     unmist
    0.91
     littered
    0.91
     reminiscent
    0.90
     rewarded
    0.86
     shaping
    0.86
    Act Density 0.518%

    No Known Activations