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

    Neuronpedia

    INDEX
    Explanations

    phrases related to ensuring control and confirmation in various contexts

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

    Following personal pronouns

    np_acts-logits-general · gemini-2.0-flash

    ensure correctness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    sizeCache
    -0.66
    IndentedString
    -0.65
    sidemargin
    -0.58
     invokingState
    -0.57
    almost
    -0.57
    casi
    -0.56
    wykle
    -0.55
     arbitrarily
    -0.53
     Almost
    -0.52
    crollView
    -0.52
    POSITIVE LOGITS
     properly
    1.58
     correctly
    1.58
     correct
    1.54
     proper
    1.36
    properly
    1.30
    correct
    1.29
     PROPER
    1.27
     correctement
    1.26
    Correct
    1.24
     Properly
    1.24
    Act Density 0.372%

    No Known Activations