Changes / Feeds / Qwen/Qwen-72B

Qwen/Qwen-72B

3 changes to a tracked key since 2023-11-26, newest first. Head as last read: b8e18ac61df64d35308695769ff46b976b6a00f4.

GPU memory to fine-tune it, card by card: Qwen-72B on /m/.

Date	Advisory	What changed, and what it does
2023-11-30	x_NXMC-2023-0029	tokenizer model_max_length 8192 to 32768; truncation defaults change (2023-11-30). `443804a5f978 -> 4c539df14494`
2023-11-28	x_NXMC-2023-0028 breaking	context window 8192 to 32768 tokens (2023-11-28). `b4a48ab18799 -> 443804a5f978`
2023-11-27	x_NXMC-2023-0027 breaking	declared dtype "bfloat16" to unset (2023-11-27). `15a041f841af -> b4a48ab18799`

Pin it, and let a bot move the pin

from_pretrained("Qwen/Qwen-72B", revision="b8e18ac61df64d35308695769ff46b976b6a00f4")

With the Renovate preset, a pull request moves this pin when the head moves, and links back here.

config changes badge

[![config changes](https://img.shields.io/endpoint?url=https%3A%2F%2Fwww.neurarch.com%2Fchanges-feeds%2Fbadge%2FQwen%2FQwen-72B.json)](https://www.neurarch.com/changes-feeds/models/Qwen/Qwen-72B.html)

Renovate datasource Badge JSON On Hugging Face Memory and fit