Grigor Nalbandyan

Hi @VainF did you look at this issue?

I am using 'yahma/llama-7b-hf'

With the checkpoint you specified, I could replicate the metrics. Do you know what is the difference between those 2? I thought there is one LLama and the checkpoints should...

I checked both the model and the tokenizer. Model weights and tokenizer.get_vocab() are the same, but there is the difference of special tokens - for baffo32 all three special tokens...