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

    Neuronpedia

    INDEX
    Explanations

    specific terms related to mathematical frameworks and operations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ptune
    -0.15
     Barton
    -0.15
    èĪ
    -0.15
    ungan
    -0.14
    untos
    -0.14
    lichkeit
    -0.14
    ilim
    -0.14
    æĥij
    -0.14
    eut
    -0.14
    addock
    -0.14
    POSITIVE LOGITS
     sp
    0.26
     sparse
    0.25
     recovery
    0.23
     compressed
    0.23
     reconstruction
    0.23
     dictionary
    0.22
     signal
    0.22
     Recovery
    0.21
     compress
    0.21
     Dictionary
    0.21
    Act Density 0.017%

    No Known Activations