I almost managed to get streaming to work against your API after some reverse-engineering, but then ran into a show-stopper bug where incomplete tokens were returned in a way that means I can't stream without accidentally displaying them
COMPUTING
-
Memory and Battery Limitations in Laptop Development
By
–
i find memory and battery to be a limiting factor for laptop development. i'm still stuck on an M1 macbook with 16GB memory, and once i start doing computing on the laptop battery depletes too quickly (like on a flight and stuff).
of course i can buy one of those ultra macbooks, -
PyTorch DDP Gradient Syncing Bug in Distributed Training
By
–
so last week I was posting about a bug i had with gradient syncing in torch DistributedDataParallel (DDP), aka the most standard way to do multi-GPU training for neural networks all of my misery originated with a single design decision from the torch team: DDP shares
-
Enterprise GenAI Training and Scaling Strategies with Cerebras
By
–
Cerebras VP and Field CTO, Natalia Vassilieva, speaks at @wandb Fully Connected Conference. Natalia will discuss training recipes and scaling strategies for high-quality enterprise GenAI Models. April 18th, 2024 3:50 PM – 4:15 PM PDT Registration details –
-
Modeling Different Coprocessors with Varying Instruction Sets
By
–
another layer of the onion 🙂 maybe can you model it like having different kinds of coprocessors with different instruction sets / reliability rates
-
Visualizing Computation DAGs: Identifying Narrow and Wide Bottlenecks
By
–
sorry i just made that word up in my head right now, didn't mean to hijack some existing term. I meant – imagine the computation as a DAG, lay it out in your head, and look for "narrow" and "wide" parts. Open to alternatives for future use!
-
Computing History Repeats: From Bytes to Tokens
By
–
The history of computing is repeating in an echo, except replace computers that do precise arithmetic on bytes with computers that do statistical arithmetic on tokens.
-
C Language Overhead: Assembly Alternative Discussion
By
–
not C. too much overhead. try assembly instead
-
Scheduling Computational Workloads to Run on Humans
By
–
# scheduling workloads to run on humans Some computational workloads in human organizations are best "run on a CPU": take one single, highly competent person and assign them a task to complete in a single-threaded fashion, without synchronization. Usually the best fit when
-
Tax Technology Automation Reduces Crypto Data Processing by 99%
By
–
Some call it the Wild West, trying to determine tax implications of #crypto derived income. #TaxTechnology teams at @RSM_Global used to spend weeks manually decoding transactions. Find out how they have improved #DataProcessing time by 99%: https://
ow.ly/SL5950RfshA #TaxData