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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating potential or hypothetical scenarios

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

    Follows "can", "could", or "can't"

    np_acts-logits-general · gemini-2.0-flash

    can be said

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    HomeAsUp
    -0.46
    からです
    -0.43
    ValueGeneration
    -0.41
    ghold
    -0.41
    віду
    -0.41
    Bre
    -0.40
    tilhe
    -0.40
    不通
    -0.40
    听
    -0.40
    र्फ
    -0.40
    POSITIVE LOGITS
     say
    1.47
    say
    1.19
     saying
    1.18
     decir
    1.13
     Saying
    1.11
    Say
    1.08
    Saying
    1.07
     SAY
    1.06
     сказать
    1.05
    saying
    1.05
    Act Density 0.316%

    No Known Activations