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

    Neuronpedia

    INDEX
    Explanations

    contrastive conjunctions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     zudem
    0.38
     daher
    0.37
    .<
    0.36
     latter
    0.36
     而且
    0.36
    #%%
    0.36
     néanmoins
    0.36
     außerdem
    0.35
     posibles
    0.35
    但是我
    0.35
    POSITIVE LOGITS
     Unlike
    0.74
     Although
    0.64
     Despite
    0.63
      
    0.63
    Unlike
    0.61
     Instead
    0.61
     While
    0.60
     Throughout
    0.60
     Like
    0.60
     Because
    0.58
    Act Density 1.265%

    No Known Activations