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

    Neuronpedia

    INDEX
    Explanations

    adjectives that describe quality and emotional resonance

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

    Appearing before descriptive adjectives

    np_acts-logits-general · gemini-2.0-flash

    positive amplification

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     виправивши
    -0.49
    #![
    -0.45
    ⁦
    -0.40
     necessarily
    -0.40
    ressee
    -0.39
    Personendaten
    -0.39
    CrossRef
    -0.37
    InputTagHelper
    -0.37
     frantically
    -0.36
     getID
    -0.36
    POSITIVE LOGITS
     ביותר
    0.76
     enough
    0.76
    ly
    0.71
    enough
    0.71
     Enough
    0.68
     للغاية
    0.66
    looking
    0.64
     yet
    0.62
     ترین
    0.60
     ENOUGH
    0.60
    Act Density 0.905%

    No Known Activations