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

    Neuronpedia

    INDEX
    Explanations

    tokenization, tokens, and tokenizers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     стены
    0.46
     Utt
    0.41
     пово
    0.41
    Utt
    0.38
     기억
    0.37
     Йо
    0.37
    borderwidth
    0.36
    信する
    0.36
     dimas
    0.35
    issait
    0.35
    POSITIVE LOGITS
     analyzers
    0.42
     tokenizer
    0.41
     wreckage
    0.40
     reglas
    0.40
     algorit
    0.40
     NuGet
    0.40
     watershed
    0.39
    teenth
    0.38
     rules
    0.38
    Piece
    0.38
    Act Density 0.019%

    No Known Activations