8Stratum: System-Hardware Co-Design with 3D-Stackable DRAM for Efficient MoeFor those confused by the headline, HN changed it from MoE to Moe.Though, thinking about "efficient moe" I'm fondly reminded of projects like https://make.girls.moe/ (2017) - "Towards the Automatic Anime Characters Creation with Generative Adversarial Networks" [sic] https://arxiv.org/abs/1708.05509A simpler time, for sure.Now that you mentioned it, I’m thinking about the Three StoogesOr Billy Idol: “In the midnight hour, she cried Moe, Moe, Moe”Seems to me similar hardware with 3D-stackable RAM could be very useful also for more "traditional" workloads.V-Cache is pretty great.
For those confused by the headline, HN changed it from MoE to Moe.Though, thinking about "efficient moe" I'm fondly reminded of projects like https://make.girls.moe/ (2017) - "Towards the Automatic Anime Characters Creation with Generative Adversarial Networks" [sic] https://arxiv.org/abs/1708.05509A simpler time, for sure.Now that you mentioned it, I’m thinking about the Three StoogesOr Billy Idol: “In the midnight hour, she cried Moe, Moe, Moe”
Now that you mentioned it, I’m thinking about the Three StoogesOr Billy Idol: “In the midnight hour, she cried Moe, Moe, Moe”
Seems to me similar hardware with 3D-stackable RAM could be very useful also for more "traditional" workloads.V-Cache is pretty great.
For those confused by the headline, HN changed it from MoE to Moe.
Though, thinking about "efficient moe" I'm fondly reminded of projects like https://make.girls.moe/ (2017) - "Towards the Automatic Anime Characters Creation with Generative Adversarial Networks" [sic] https://arxiv.org/abs/1708.05509
A simpler time, for sure.
Now that you mentioned it, I’m thinking about the Three Stooges
Or Billy Idol: “In the midnight hour, she cried Moe, Moe, Moe”
Seems to me similar hardware with 3D-stackable RAM could be very useful also for more "traditional" workloads.
V-Cache is pretty great.