Nguyễn Lê Phúc Vinh

Con người "mê" code =)))))

Results 5 issues of


                                            Nguyễn Lê Phúc Vinh

Training is ineffective, the generator's loss is only around 5,000

Training is ineffective, the generator's loss is only around 5,000 but discriminator's loss is 0.34

Compare MobiLLama and LLM360's Amber?

What reflow sound like?, any sample for reflow?

2

comment

Xin phép được sử dụng tên "racoon" trong project cá nhân.

binarize the dataset and save to a file like: train.bin, val.bin instead of re binarize dataset from scratch