Mark Schmidt
Mark Schmidt
@hahayusuf As someone pointed out in the PR, some countries do not have access to the OpenAI API directly and must use proxies. An option to set this variable in...
Like @olliestanley said, the license is the license. Any ethical document/guide/recommendations doesn't have a place here. > No restrictions beyond the Apache 2.0 License are or will be imposed by...
It's not a leak, this is just an artifact of how language models work. These responses are a mix of training data and hallucinations not chat leaks. You can get...
@TavernAI https://github.com/TavernAI/TavernAI/issues/1 appears incorrectly closed. Anyway, here is a working fork of TavernAI with Horde support: https://github.com/TavernAI/TavernAI/compare/main...Aspartame-e951:TavernAI:main
Okay, I made the issue title less alarming since you've chimed in. Open communication about the issue and what is being done to address it would be appreciated by many....
There's a 4.9GB [ggml 4bit GPTQ quantization](https://huggingface.co/cakewalk/ggml-q4_0-stablelm-tuned-alpha-7b/tree/main) for StableLM-7B up on HuggingFace which works in llama.cpp for fast CPU inference. (For comparison, LLaMA-7B in the same format is 4.1GB. But,...
This is an early Alpha training checkpoint. It is going to be rough around the edges until training finishes.
Language Models give random outputs. To get the exact output in their ReadMe you would have to know what random seed lead to that output and then set the same...
It's still using the OpenAI API, just not their endpoint, even if the model behind it isn't an OpenAI model.