@gpusteve: you're interviewing for an ml performance role at anthropic and they ask: "you're serving a 70b transformer model on 8 …

X AI KOLs Timeline News

Summary

A tweet explains the correct answer to an ML performance interview question at Anthropic about the latency tradeoffs of splitting tensor-parallel linear layers by columns vs. rows when serving a 70B transformer model on 8 GPUs, highlighting that performance is not similar despite equal per-GPU weights.

you're interviewing for an ml performance role at anthropic and they ask: "you're serving a 70b transformer model on 8 gpus. what are the latency tradeoffs of splitting tensor-parallel linear layers by columns vs rows?" you say: "either way each gpu gets 1/8 of the weights, so performance should be similar." wrong. here's how you answer:
Original Article
View Cached Full Text

Cached at: 06/08/26, 05:19 AM

you’re interviewing for an ml performance role at anthropic and they ask:

“you’re serving a 70b transformer model on 8 gpus. what are the latency tradeoffs of splitting tensor-parallel linear layers by columns vs rows?”

you say: “either way each gpu gets 1/8 of the weights, so performance should be similar.” wrong.

here’s how you answer:

start with what the linear layer is doing.

it takes the current hidden states X and multiplies them by a weight matrix W to produce Y.

tensor parallelism shards W across gpus. the important detail is which dimension you shard, because that decides the shape of the partial outputs.

real

real

Similar Articles