Changes / Feeds / Qwen/Qwen2.5-3B
Qwen/Qwen2.5-3B
3 changes to a tracked key since 2024-09-15, newest first. Head as last read: 3aab1f1954e9cc14eb9509a215f9e5ca08227a9b.
| Date | Advisory | What changed, and what it does |
|---|---|---|
| 2024-09-20 | x_NXMC-2024-0259 breaking | EOS changed from "<|im_end|>" to "<|endoftext|>"; anything fine-tuned or served against the old value stops at a different token (2024-09-20).e4aa5ac50aa5 -> 3aab1f1954e9 Open in app |
| 2024-09-19 | x_NXMC-2024-0258 breaking | the chat template changed and renders differently (tool-call, tool-call-parts); prompts rendered with the old one no longer match what the model now expects, so cached prompts and eval sets need re-rendering (2024-09-19).c10dcf3d5e82 -> e4aa5ac50aa5 Open in app |
| 2024-09-16 | x_NXMC-2024-0257 breaking | the chat template changed and renders differently (tool-call, tool-call-parts); prompts rendered with the old one no longer match what the model now expects, so cached prompts and eval sets need re-rendering (2024-09-16).a895be0c08c4 -> c10dcf3d5e82 Open in app |
Pin it, and let a bot move the pin
from_pretrained("Qwen/Qwen2.5-3B", revision="3aab1f1954e9cc14eb9509a215f9e5ca08227a9b")
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/Qwen/Qwen2.5-3B.html)