INDEX
    Explanations

    high-level topic or entity labels—nouns that define the kind of project, platform, or organization being described, especially in titles or user prompts.

    New Auto-Interp
    Negative Logits
     manchmal
    0.46
     intramolecular
    0.43
     normalerweise
    0.41
     Gemä
    0.39
     selten
    0.38
     begrij
    0.37
     பழைய
    0.37
     discretized
    0.37
     Sekunden
    0.37
     erreur
    0.36
    POSITIVE LOGITS
     startup
    0.64
     설립
    0.56
     enterprise
    0.55
     startups
    0.54
     business
    0.54
    business
    0.53
    成立
    0.53
     establishment
    0.53
     launch
    0.52
     gegründet
    0.52
    Act Density 0.226%

    No Known Activations