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

    Neuronpedia

    INDEX
    Explanations

    occurrences of the conjunction "and" and related terms within the context

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

    the word "and" connecting clauses

    np_acts-logits-general · gemini-2.0-flash

    and then a specific word

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     obstante
    -0.56
     mérit
    -0.45
     CreateTagHelper
    -0.45
     embora
    -0.44
     pertenecen
    -0.43
     Estaba
    -0.43
     sarebbero
    -0.42
    følgelig
    -0.41
    ostante
    -0.41
    tvrt
    -0.40
    POSITIVE LOGITS
     introduce
    1.05
     prepare
    0.99
     took
    0.94
     make
    0.92
     allocate
    0.89
     get
    0.88
     apply
    0.86
     undertake
    0.86
     remove
    0.85
     decide
    0.85
    Act Density 1.132%

    No Known Activations