← Back to context Comment by Tejush 1 hour ago Any guess on the pre-training tokens/flops they consumed? 1 comment Tejush Reply lostmsu 15 minutes ago Technical report has a graph vs Kimi K2 with 1e21 FLOPs (but they don't claim that's the entirety of pretraining)
lostmsu 15 minutes ago Technical report has a graph vs Kimi K2 with 1e21 FLOPs (but they don't claim that's the entirety of pretraining)
Technical report has a graph vs Kimi K2 with 1e21 FLOPs (but they don't claim that's the entirety of pretraining)