erichan1 issues

Repositories
Issues
Comments

Results 2 issues of


                                            erichan1

llm int8 integration into Megatron part of metaseq

Add LLM int8 to Megatron. Steps to use 1. Flip on QUANTIZED_INFERENCE in layers.py 2. Comment out _log_weight_stats [here](https://github.com/facebookresearch/metaseq/blob/905f355e3ddadaf9e50f0179006f2295917cfbda/metaseq/models/transformer_decoder.py#L241). Not sure if this is simply buggy or that the weight...

Cannot index into a NestedTensor with grad ie nt[0]

## 🐛 Bug Can index into a NestedTensor without grad, but cannot index into a NestedTensor with grad ## To Reproduce Steps to reproduce the behavior: ``` import torch c...