egghead VS test

It's a toy I threw together as a weekend project, but you're welcome to give it a whirl: https://github.com/toasterrepairman/egghead

Here's the rundown:

- You need libtorch, openssl and cargo installed on your system before compiling

AND

- You have to put the variables from the README in your ~.bashrc along with a valid Discord bot token

Once you do that, it should "just work". It's using a super pruned model with high-temperature tuning, so the results should be... dicy. I assume no responsibility for the vast amount of misinformation this will produce.

Commands include "e.help" for help, "e.ask" for traditional ChatGPT-style questions, "e.news" to grab a Fox headline and generate the rest, "e.wiki" to look up a Wikipedia article and use it as a prompt, and "e.hn"... a feature I will build Soon™.

Let me know if you run into any issues!

Sources [1] MMLU Benchmark (Multi-task Language Understanding) | Papers With Code https://paperswithcode.com/sota/multi-task-language-understanding-on-mmlu [2] MMLU Dataset | Papers With Code https://paperswithcode.com/dataset/mmlu [3] hendrycks/test: Measuring Massive Multitask Language Understanding | ICLR 2021 - GitHub https://github.com/hendrycks/test [4] lukaemon/mmlu · Datasets at Hugging Face https://huggingface.co/datasets/lukaemon/mmlu [5] [2009.03300] Measuring Massive Multitask Language Understanding - arXiv https://arxiv.org/abs/2009.03300

With a score of 90.0%, Gemini Ultra is the first model to outperform human experts on MMLU (massive multitask language understanding), which uses a combination of 57 subjects such as math, physics, history, law, medicine and ethics for testing both world knowledge and problem-solving abilities.

One method for comparison is the MMLU https://arxiv.org/abs/2009.03300.

Because there are many benchmarks that measure different things.

You need to look at the benchmark that reflects your specific interest.

So in this case ("I wasn't impressed that 30B didn't seem to know who Captain Picard was") the closest relevant benchmark they performed is MMLU (Massive Multitask Language Understanding"[1].

In the LLAMA paper they publish a figure of 63.4% for the 5-shot average setting without fine tuning on the 65B model, and 68.9% after fine tuning. This is significantly better that the original GPT-3 (43.9% under the same conditions) but as they note:

> "[it is] still far from the state-of-the-art, that is 77.4 for GPT code-davinci-002 on MMLU (numbers taken from Iyer et al. (2022))"

InstructGPT[2] (which OpenAI points at as most relevant ChatGPT publication) doesn't report MMLU performance.

[1] https://github.com/hendrycks/test

[2] https://arxiv.org/abs/2203.02155

Benchmark result is 67.6% which is 7.6% improvement from Gopher. MMLU is multiple choice Q&A over various subjects. Questions can be found linked in this github repo (see data).