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

    Neuronpedia

    INDEX
    Explanations

    hallucinations and LLM properties

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     portátil
    0.48
     Swarovski
    0.43
     cargos
    0.43
     nascimento
    0.43
    achta
    0.43
     líquido
    0.42
    vať
    0.42
     neodymium
    0.41
    鐲
    0.41
     vendidos
    0.41
    POSITIVE LOGITS
     describing
    0.47
    History
    0.43
    Needed
    0.42
     HISTORY
    0.42
    Ghost
    0.42
    ظام
    0.41
    Comple
    0.41
     protects
    0.41
     chapter
    0.40
     meaningfully
    0.40
    Act Density 0.001%

    No Known Activations