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

    Neuronpedia

    INDEX
    Explanations

    expressions of preference or desire for alternatives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    preferring one thing over another

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    nezeu
    -0.51
    AddHtmlAttribute
    -0.50
     transfieras
    -0.50
     iſt
    -0.48
     nonUne
    -0.47
     canst
    -0.45
    StructEnd
    -0.45
     ſever
    -0.44
     againſt
    -0.44
    ToScroll
    -0.44
    POSITIVE LOGITS
     liever
    0.92
     prefer
    0.84
     prefers
    0.79
     prefier
    0.78
     preferring
    0.78
     rather
    0.77
     preferred
    0.76
     préf
    0.75
     prefi
    0.74
    preferred
    0.73
    Act Density 0.039%

    No Known Activations