Sergey Sosnin

Is it ok to use leaked LLaMA for research?

2

I would like to pose a question: Is it appropriate for the scientific community to utilize LLaMA for research if the application has not been explicitly approved? This inquiry seems...

Ray parallelization does not work

1

**Describe the bug** Does not work model parallelization with Ray and a custom model from huggingface. **To Reproduce** I want to train a neural network using ludwig and molecular encoder...

Impossibility to use a tokenizer with auto_transformer

4

I want to use [this model](https://huggingface.co/ibm/MoLFormer-XL-both-10pct) as an encoder. As you can see from the description, the model can be uploaded like: ``` model = AutoModel.from_pretrained("ibm/MoLFormer-XL-both-10pct", deterministic_eval=True, trust_remote_code=True) tokenizer =...

Update transformers.py

1

fix numpy int error -> int32

saving a model into HDF5

7

Hi, Is it possible to dump a model into HDF5 file? The problem is when i try to dump 1500 samples with your function save_prefix the total disk space required...