PriNova

> the tokenizer.model contains scores for each token, most of which are just the negation of the token index (since they're output by the bpe trainer in descending order) so...

And the user do not need to change the code for defining the right part count of the model. With the option `--n_parts` can be set to the number of...

> I noticed that often the interactive mode (used as a chat with for example the `chat-with-bob.txt` initial prompt) fails due to **LLaMA trying to escape** the chat (mainly with...

> @PriNova so you didn't have the issue before #252? Have you tried using the older version of the repo to see if the issue disappears? No, it works like...

> Hi everyone, I hacked together a python script to convert a model saved as GGML Q4_0 files back to Pytorch checkpoint for further consumption/training/finetuning using HuggingFace's Transformer package and/or...

Wow, Fantastic. Thank you for this contribution.

> Hello, > > I noticed something when trying the chat with Bob is that I always get the first token as empty. > > ``` > 1 -> ''...

> What you're talking about is training/finetuning which is theoretically possible on CPU but practically impossible/non-feasible on CPU only because you'll be training for literal months instead of days, you...

You get different output text because of the temperature setting. To really get deterministic outputs you have to set the temperature nearly to zero (but not exactly zero because then...

> @PriNova Nah, when you fix the seed (I put seed = 1) you'll get the same output everytime. Try it by yourself, put a fixed seed + high temperature...