view post Post 5421 Surya-1.1T: Scaling Beyond Human-Level Reasoning via 146 Trillion Token Pre-trainingAuthor: SKT AI LABSAffiliation: SKT AI Labs / Project SuryaModel Architecture: Optimized Dense TransformerParameters: 1.1 TrillionTraining Tokens: 146 TrillionWanna collaborate us Friends let's Start Journey we have Collected 146 trillon tokens and done pre training but we need to made more powerfullWhitepaper - https://github.com/SHRIJANAGAIN/PROFF See translation 56 replies Β· π₯ 15 15 π 9 9 π 8 8 π€ 7 7 β 7 7 π 6 6 β€οΈ 6 6 π 5 5 π§ 5 5 π€ 5 5 π€― 3 3 + Reply