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

    Neuronpedia

    INDEX
    Explanations

    code-related constructs involving conditions and comparisons

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

    mathematical symbols or code operators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     queſta
    -1.10
    ⟬
    -1.05
    <unused79>
    -0.97
    <unused68>
    -0.97
    <unused43>
    -0.96
    <unused41>
    -0.96
    <unused51>
    -0.96
    <unused52>
    -0.96
    <unused8>
    -0.96
    <unused28>
    -0.96
    POSITIVE LOGITS
    ↵↵
    0.50
     hacerlo
    0.50
     equally
    0.42
    ↵
    0.40
     equivalent
    0.39
    <eos>
    0.39
     likewise
    0.38
     farlo
    0.38
     same
    0.38
     both
    0.37
    Act Density 0.041%

    No Known Activations