Tag: 显存优化 (2 articles)

Multiverse Computing introduces a memory-efficient knowledge distillation technique that uses cached top-K logits and a fused chunked KL loss to train on long contexts with just a single GPU, drastically reducing costs.

NVIDIA NeMo AutoModel seamlessly plugs into the HuggingFace ecosystem, boosting MoE fine-tuning throughput by 3.4x-3.7x and cutting VRAM usage by 30% with a single import line change.