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

    Neuronpedia

    INDEX
    Explanations

    contrastive conjunctions and transitions indicating a shift or addition in discussion

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    __*/
    -0.52
        
    -0.47
     hochzeit
    -0.43
    vací
    -0.43
    WithIOException
    -0.43
      
    -0.43
     kjø
    -0.43
    emlrt
    -0.42
     fofo
    -0.42
    ":[{
    -0.42
    POSITIVE LOGITS
    Nhưng
    0.86
     Bourgoin
    0.82
    though
    0.81
    Tetapi
    0.81
     Zwar
    0.81
     ]
    0.80
     مشين
    0.80
     تضيفلها
    0.79
     nevertheless
    0.77
     ་་
    0.76
    Act Density 0.353%

    No Known Activations