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

    Neuronpedia

    INDEX
    Explanations

    expressions indicating improvement or enhancement

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

    better + verb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     pinulongan
    -0.59
     avoient
    -0.54
     étoient
    -0.52
     hvě
    -0.46
     Turquía
    -0.46
     propOrder
    -0.45
    makaian
    -0.45
     casera
    -0.45
     auroit
    -0.44
     coû
    -0.43
    POSITIVE LOGITS
     understand
    0.63
     describe
    0.58
     understanding
    0.57
     support
    0.56
     define
    0.54
    tize
    0.52
    segue
    0.52
    不易
    0.52
    bien
    0.52
     describes
    0.52
    Act Density 0.017%

    No Known Activations