InferenceIllusionist / dolphin-2.9.1-mixtral-1x22b-iMat-GGUF
Other1.7K0Updated May 24, 2024
Downloads
1.7K
Likes
0
Last updated
May 24, 2024
License
apache-2.0
Quantized from fp16. Weighted quantizations were creating using fp16 GGUF and groupsmerged-enhancedV2-TurboMini.txt in 228 chunks and nctx=512 This method of calculating the importance matrix showed improvements in some areas for Mistral 7b and Llama3 8b models, see above post…
Tasks & library
- transformers
Tags
- transformers
- gguf
- generated_from_trainer
- axolotl
- iMat
- en
Technical specifications
ParametersNot available
Librarytransformers
Languagesen
Licenseapache-2.0