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

    Neuronpedia

    INDEX
    Explanations

    references to various types of "benches" in different contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     marito
    -0.72
     soeur
    -0.69
     sœur
    -0.68
    voerd
    -0.65
    hasMoreElements
    -0.64
     rempliss
    -0.64
     sonno
    -0.63
     réparation
    -0.63
     dragen
    -0.63
     touristes
    -0.63
    POSITIVE LOGITS
     bench
    3.39
     Bench
    3.16
    Bench
    2.99
    bench
    2.89
     benches
    2.55
    BENCH
    2.44
     Benchmark
    1.96
     benchmark
    1.85
     benchmarks
    1.79
    Benchmark
    1.77
    Act Density 0.088%

    No Known Activations