gpt-2 VS metaseq

Medium models: Roughly between 1B to 10B parameters. This is where Mistral 7B, Phi-3, Gemma from Google DeepMind, and wizardlm2 sit. Fun fact: GPT 2 was a medium sized model, much smaller than its latest versions.

"Our model, called GPT-2 (a successor to GPT), was trained simply to predict the next word in 40GB of Internet text. Due to our concerns about malicious applications of the technology, we are not releasing the trained model. As an experiment in responsible disclosure, we are instead releasing a much smaller model for researchers to experiment with, as well as a technical paper. "

They did release GPT-2 under the MIT License.

Did you arrive at this certainty through reading something other than what OpenAI has published? The document [0] that describes the training data for GPT-2 makes this assertion hilarious to me.

[0]: https://github.com/openai/gpt-2/blob/master/model_card.md#da...

I made a clone of https://github.com/openai/gpt-2 on my local laptop

This is a super important issue that affects the pace and breadth of iteration of AI almost as much as the raw hardware improvements do. The blog is fun but somewhat shallow and not technical or very surprising if you’ve worked with clusters of GPUs in any capacity over the years. (I liked the perspective of a former googler, but I’m not sure why past colleagues would recommend Jax over pytorch for LLMs outside of Google.) I hope this newco eventually releases a more technical report about their training adventures, like the PDF file here: https://github.com/facebookresearch/metaseq/tree/main/projec...

The number of people who can actually pre-train a true LLM is very small.

It remains a major feat with many tweaks and tricks. Case in point: the 114 pages of OPT175B logbook [1]

[1] https://github.com/facebookresearch/metaseq/blob/main/projec...

Reliability at scale: take a look at the OPT training log book for their 175B model run. It needed a lot of babysitting. In my experience, that scale of TPU training run requires a restart about once every 1-2 weeks—and they provide the middleware to monitor the health of the cluster and pick up on hardware failures.

I really appreciated this log of Facebook training a large language model of how troublesome AI development can be: https://github.com/facebookresearch/metaseq/tree/main/projects/OPT/chronicles

Stable Diffusion will run on any decent gaming GPU or a modern MacBook, meanwhile LLMs comparable to GPT-3/ChatGPT have had pretty insane memory requirements - e.g., <https://github.com/facebookresearch/metaseq/issues/146>

It seems so, check this log book from Meta: https://github.com/facebookresearch/metaseq/blob/main/projec...