File size: 983 Bytes
1bcfe7d 098e39b 1bcfe7d 098e39b cb587b3 fa87de1 128e0a0 fa87de1 b6e21b4 21d9779 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
---
license: other
license_name: yi-license
license_link: https://huggingface.co/01-ai/Yi-34B-200K/blob/main/LICENSE
pipeline_tag: text-generation
library_name: gguf
---
GGUF importance matrix (imatrix) quants for https://huggingface.co/jondurbin/bagel-34b-v0.4
The importance matrix was trained for ~50K tokens (105 batches of 512 tokens) using a [general purpose imatrix calibration dataset](https://github.com/ggerganov/llama.cpp/discussions/5263#discussioncomment-8395384).
Prompt strategies: https://huggingface.co/jondurbin/bagel-34b-v0.4#prompting-strategies
Feb. 21, 2024: Updating quants from [latest commit](https://huggingface.co/jondurbin/bagel-34b-v0.4/commit/94054270c89880c5fbc7e8d9d7b7540fcfdcbbeb).
| Layers | Context | [Template](https://huggingface.co/jondurbin/bagel-34b-v0.4#prompt-formatting) |
| --- | --- | --- |
| <pre>60</pre> | <pre>200000</pre> | <pre>[INST] \<\<SYS\>\><br>{instructions}<br>\<\</SYS\>\><br><br>{prompt} [/INST]<br>{response}</pre> | |