YaLM-100B
YaLM-100B copied to clipboard
yandex
Pretrained language model with 100B parameters
`generate_*.sh` scripts appear to create `fp16` model and load weights into it, but the weights are `bfloat16`.
Sorry for maybe a stupid question, fortunately, I find your product and want to integrate it into social media accounts; I don't release how to use it from the box...
Please verify that checksums in the torrent file match before merging this.
It's great that you endure the community. Thank you. Please add an online example so that you can test with it without downloading 200GB to your computer.
Thanks for open-sourcing this! Because the GPU ram requirements are so high, it's hard to rent a large enough single node from any of the major cloud providers. How can...
This model looks amazing, thank you! We have a machine with 8 x 3090 (192GB total), I tried to run the examples, but I get: ``` building GPT2 model ......
Thanks for the awesome work! (and a especially for choosing to make it freely available) If you have time, please also consider running the evaluation benchmarks from lm-eval-harness https://github.com/EleutherAI/lm-evaluation-harness [despite...
