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

    Neuronpedia

    INDEX
    Explanations

    information related to parameters and thresholds in various contexts

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

    Values, levels, thresholds, or concentrations

    np_acts-logits-general · gemini-2.0-flash

    measures or quantities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Names
    -0.74
     Sizes
    -0.73
     Ratios
    -0.73
     genders
    -0.72
     Prices
    -0.72
     Grades
    -0.71
     Concentrations
    -0.70
     Ages
    -0.70
    Names
    -0.70
     Preisen
    -0.69
    POSITIVE LOGITS
     value
    1.11
     score
    0.98
     rate
    0.94
     price
    0.89
     figure
    0.84
     temperature
    0.76
     speed
    0.72
     height
    0.70
     weight
    0.70
     size
    0.68
    Act Density 1.277%

    No Known Activations