Pretraining From Scratch on an Inference-Only NPU: VKORE-46M-NPU
DEV Community
Pretraining From Scratch on an Inference-Only NPU: VKORE-46M-NPU
A language model pretrained from scratch on an inference-only NPU via real backpropagation, released on Hugging Face. Loss trajectory correlates 0.99 with a GPU reference.
0 comments
No comments yet.