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

    Neuronpedia

    INDEX
    Explanations

    concepts related to deep learning feature extraction and evaluation metrics

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

    Features of trained CNNs

    np_acts-logits-general · gemini-2.0-flash

    trained detectors and descriptors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     idén
    -0.57
    InvalidProtocol
    -0.56
     AssemblyVersion
    -0.54
    archiviato
    -0.53
    êque
    -0.52
    ragment
    -0.52
     idéia
    -0.50
     refusé
    -0.49
     swatches
    -0.48
    SourceChecksum
    -0.47
    POSITIVE LOGITS
    adaptiveStyles
    0.83
     machine
    0.69
     Machine
    0.67
     disent
    0.66
     interpre
    0.66
     deep
    0.63
     Transfer
    0.62
     transfer
    0.61
    Machine
    0.59
     adversarial
    0.59
    Act Density 1.384%

    No Known Activations