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

    Neuronpedia

    INDEX
    Explanations

    references to research methodologies and scientific processes

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

    phrases related to research methods, statistical analysis, and technical details of scientific studies.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    This neuron detects prepositions (especially "in") when they appear in academic or technical contexts describing relationships between concepts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Negative Logits
     autorytatywna
    -0.85
    AnimationsModule
    -0.70
    principalTable
    -0.63
     Italijanski
    -0.62
     Мексичка
    -0.62
     виправивши
    -0.60
    <unused52>
    -0.59
    <unused47>
    -0.58
    [@BOS@]
    -0.58
     Waſſer
    -0.58
    POSITIVE LOGITS
     both
    0.39
     elsewhere
    0.38
     emprego
    0.38
    tatuajes
    0.38
     både
    0.37
     nocturna
    0.36
     vieill
    0.35
     vété
    0.35
     bý
    0.35
     throughout
    0.35
    Act Density 0.553%

    No Known Activations