• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI
    Reaction

    Sol-scale models could match Bel’s math performance within six months

    A user suspects Bel was trained on extensive synthetic data involving complex math problems.

    Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)T(
    1 Source, 46m ago, first seen 46m ago

    TLDR

    A user suspects Bel was trained on extensive synthetic data involving complex math problems. They predict Sol-scale models, or even smaller ones, could match Bel’s math performance within six months because of data.

    Combined views

    1.2K

    1 Source, first seen 46m ago

    Combined views

    1.2K

    1 Source, first seen 46m ago

    10 likes
    10 likes
    1 comments
    3 saves
    1 comments
    3 saves

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    1 Source

    Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTexI doubt this and expect that the main difference is that Bel was trained on a ton of synthetic data covering nearly as complex math problems, and in 6 months we'll see equal math performance from Sol-scale models if not smaller, again due to data.2h

    1 Source

    Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTexI doubt this and expect that the main difference is that Bel was trained on a ton of synthetic data covering nearly as complex math problems, and in 6 months we'll see equal math performance from Sol-scale models if not smaller, again due to data.2h

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.