shaileshj2803

Results 6 comments of shaileshj2803

Have a sdist would be better. If you can put it in a backlog that will really help us.

I was able to train using deepspeed on 8 V100 GPUs. Here is the training script and deepseed config file. torchrun --nproc_per_node=8 --master_port=9776 train.py \ --model_name_or_path hf_model/llama-7b \ --data_path ./alpaca_data.json...

i have the same question. Do you have plans for openllama2?

Can we please have a fix for this?

Can we please support python 3.10

please add support for java