KoboldAI-Client VS transformers

You can use KoboldAI to run a LLM locally. There are hundreds / thousands of models on hugging face. Some uncensored ones are Pygmalion AI (chatbot), Erebus (story writing AI), or Vicuna (general purpose).

GPT-J 6B is a pretty weak and outdated model. Nerys 13B would probably give you better replies but they lean more towards SFW stuff. Erebus was their best model for erotic roleplay but they removed it as it went against Google's TOS. You can check out their documentation here.

If you do have that kind of hardware, the next step would be looking for what model to run. I came across Kobold's models. Their main github page is here: https://github.com/KoboldAI/KoboldAI-Client

Then there are graphical user interfaces like text-generation-webui and gpt4all for general purpose chat. There are also KoboldAI and SillyTavern, they have focus more on storytelling and roleplay and have tools to improve that.

And finally the folks from the KoboldAi do some interesting stuff with Pseudocode and Soft-Prompts that might also be relevant.

Click this link. ---> https://github.com/KoboldAI/KoboldAI-Client/tree/main

Do you believe the problem could be that my KoboldAI is outdated? I did download the one from henk717 at https://github.com/KoboldAI/KoboldAI-Client but it was a little while ago.

Is t5x an encoder/decoder architecture?

Some more general options.

The Flax ecosystem

https://github.com/google/flax?tab=readme-ov-file

or dm-haiku

https://github.com/google-deepmind/dm-haiku

were some of the best developed communities in the Jax AI field

Perhaps the “trax” repo? https://github.com/google/trax

Some HF examples https://github.com/huggingface/transformers/tree/main/exampl...

Sadly it seems much of the work is proprietary these days, but one example could be Grok-1, if you customize the details. https://github.com/xai-org/grok-1/blob/main/run.py

The HuggingFace transformers library already has support for a similar method called prompt lookup decoding that uses the existing context to generate an ngram model: https://github.com/huggingface/transformers/issues/27722

I don't think it would be that hard to switch it out for a pretrained ngram model.

Most of this tutorial is based on Hugging Face course about Transformers and on Niels Rogge's Transformers tutorials: make sure to check their work and give them a star on GitHub, if you please ❤️

* Superconvergence + LR range finder + Fast AI's Ranger21 optimizer was the goto optimizer for CNNs, and worked fabulously well, but on transformers, the learning rate range finder sadi 1e-3 was the best, whilst 1e-5 was better. However, the 1 cycle learning rate stuck. https://github.com/huggingface/transformers/issues/16013

Thanks! :) I'm pushing them into transformers, pytorch-gemma and collabing with the Gemma team to resolve all the issues :)

The RoPE fix should already be in transformers 4.38.2: https://github.com/huggingface/transformers/pull/29285

My main PR for transformers which fixes most of the issues (some still left): https://github.com/huggingface/transformers/pull/29402

If you want to tinker with the architecture Hugging Face has a FOSS implementation in transformers: https://github.com/huggingface/transformers/blob/main/src/tr...

If you want to reproduce the training pipeline, you couldn't do that even if you wanted to because you don't have access to thousands of A100s.

I am aware that using mixed precision reduces the stability of weight and there will be little consistency but don't expect it to be this much. I have attached the graph of evaluation metrics. If someone can give me some insight into this issue, that would be great.