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

    Neuronpedia

    INDEX
    Explanations

    phrases that express expectations and comparisons

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

    phrases indicating comparison or deviation from expectations, especially involving contrasts like "than expected" or "as intended."

    oai_token-act-pair · gpt-4.1-2025-04-14Triggered by @ywhp2c8tjm

    Followed by a subordinate clause

    np_acts-logits-general · gemini-2.0-flash

    the way it should

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ahren
    -0.49
    EMBER
    -0.43
     Verf
    -0.40
    oo
    -0.40
    TAN
    -0.40
    odis
    -0.39
    zem
    -0.38
     کشت
    -0.38
     الترك
    -0.38
    den
    -0.38
    POSITIVE LOGITS
     متعلقه
    0.59
    IntoConstraints
    0.59
     expected
    0.57
     Majefty
    0.53
     ujednoznacz
    0.52
    expected
    0.50
     يتيمه
    0.50
     desired
    0.49
    LabelTagHelper
    0.49
     deseado
    0.49
    Act Density 0.439%

    No Known Activations