Changes / Feeds / google-t5/t5-base
google-t5/t5-base
3 changes to a tracked key since 2019-12-12, newest first. Head as last read: a9723ea7f1b39c1eae772870f3b547bf6ef7e6c1.
GPU memory to fine-tune it, card by card: t5-base on /m/.
| Date | Advisory | What changed, and what it does |
|---|---|---|
| 2022-12-12 | x_NXMC-2022-0003 breaking | architectures ["T5WithLMHeadModel"] to ["T5ForConditionalGeneration"]; a different class is instantiated from the same id (2022-12-12).c1852398fbcc -> a4986abc64c5 |
| 2020-04-24 | x_NXMC-2020-0029 breaking | EOS declared as 1 (was unset); generation that relied on the library default now stops on this token; model_type declared as "t5" (was unset); pad token unset to 0; batched training masks a different id (2020-04-24).f611a04e050c -> c1852398fbcc |
| 2020-01-31 | x_NXMC-2020-0028 | architectures declared as ["T5WithLMHeadModel"] (was unset) (2020-01-31).666f4ee1ab30 -> 06e82b573235 |
Pin it, and let a bot move the pin
from_pretrained("google-t5/t5-base", revision="a9723ea7f1b39c1eae772870f3b547bf6ef7e6c1")With the Renovate preset, a pull request moves this pin when the head moves, and links back here.
[](https://www.neurarch.com/changes-feeds/models/google-t5/t5-base.html)