Changes / Feeds / google/gemma-3-1b-it
google/gemma-3-1b-it
3 changes to a tracked key since 2025-03-12, newest first. Head as last read: dcc83ea841ab6100d6b47a070329e1ba4cf78752.
| Date | Advisory | What changed, and what it does |
|---|---|---|
| 2025-03-19 | x_NXMC-2025-0080 | generation defaults changed (do_sample unset to true, top_k unset to 64, top_p unset to 0.95); a call that does not pass these explicitly now generates differently (2025-03-19).9b99be88fdd7 -> 417a7e01fafd |
| 2025-03-12 | x_NXMC-2025-0079 breaking | EOS changed from 1 to [1,106]; anything fine-tuned or served against the old value stops at a different token (2025-03-12).ca8476f4df74 -> 9b99be88fdd7 |
| 2025-03-12 | x_NXMC-2025-0078 | the chat template changed; prompts rendered with the old one no longer match what the model now expects, so cached prompts and eval sets need re-rendering (2025-03-12).6c0ce4ee3210 -> ca8476f4df74 |
Pin it, and let a bot move the pin
from_pretrained("google/gemma-3-1b-it", revision="dcc83ea841ab6100d6b47a070329e1ba4cf78752")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/gemma-3-1b-it.html)