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

    Neuronpedia

    INDEX
    Explanations

    references to sequential instructions or questions

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

    This neuron identifies patterns where "the following" appears or where instructions/directions request completion of listed items.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    linguistic terms related to logical instructions, such as "following", "each", "whether", "according to", etc.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Negative Logits
     tartalomajánló
    -0.94
     myſelf
    -0.74
     himſelf
    -0.70
    mtliche
    -0.67
     itſelf
    -0.66
     xenia
    -0.64
     Arithmetic
    -0.61
     retrograde
    -0.61
     Eskimo
    -0.61
    OrNil
    -0.60
    POSITIVE LOGITS
     following
    0.68
    MigrationBuilder
    0.59
    CodedInputStream
    0.57
    ंदीखरीदारी
    0.56
    following
    0.56
    migrationBuilder
    0.54
    Diwedd
    0.54
    ↵
    0.52
    queryInterface
    0.51
    Filmografie
    0.50
    Act Density 0.300%

    No Known Activations