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

    Neuronpedia

    INDEX
    Explanations

    instances of transitional or contrasting phrases in the text

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

    however, yet

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     aapt
    -0.43
    posedge
    -0.38
     BoxFit
    -0.37
    utscher
    -0.36
    succeeded
    -0.36
    >>()
    -0.35
    OneToOne
    -0.35
    "}"
    -0.35
     الرياضيه
    -0.35
     Mexique
    -0.34
    POSITIVE LOGITS
     transférez
    0.56
    Obrázky
    0.52
     يتيمه
    0.45
    0.44
    WriteAttribute
    0.43
    sohn
    0.42
     scythe
    0.42
     esigenze
    0.41
    ież
    0.41
    Diweddarwch
    0.40
    Act Density 0.035%

    No Known Activations