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

    Neuronpedia

    INDEX
    Explanations

    references to "high" in various contexts, especially relating to quality, profile, or performance

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

    Follows the word "high"

    np_acts-logits-general · gemini-2.0-flash

    high things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     piernas
    -0.54
     ramai
    -0.53
     Sizes
    -0.52
    AnchorStyles
    -0.52
     prêts
    -0.51
     Size
    -0.50
     météo
    -0.50
    neath
    -0.49
     beira
    -0.49
     SIZE
    -0.48
    POSITIVE LOGITS
     octane
    0.75
    IntoConstraints
    0.73
     degree
    0.71
     profile
    0.70
    ValueStyle
    0.69
     priced
    0.68
    lighed
    0.68
     level
    0.66
    brow
    0.66
    jacking
    0.65
    Act Density 0.230%

    No Known Activations