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

    Neuronpedia

    INDEX
    Explanations

    phrases and words that indicate importance or priority in arguments

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

    Adverbs and transition words

    np_acts-logits-general · gemini-2.0-flash

    ordering or emphasis adverbs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    principalColumn
    -0.67
    tonode
    -0.65
     AssemblyCulture
    -0.63
    utilisons
    -0.62
    complexContent
    -0.61
    Diweddarwch
    -0.60
    
    -0.59
    basicConfig
    -0.59
    Бахар
    -0.56
    ällor
    -0.55
    POSITIVE LOGITS
     importantly
    1.24
    tably
    0.80
     Importantly
    0.75
     conversely
    0.73
     interestingly
    0.72
    ureusement
    0.71
     notably
    0.69
    Conversely
    0.69
    ificantly
    0.68
     moreover
    0.63
    Act Density 0.504%

    No Known Activations