CompressedGemma
/

HPC-Quantize

Model card Files Files and versions

HPC-Quantize / hexstate_requantize.py

Commit History

Q8_0 tied embeddings

f32b3c6
verified

CompressedGemma commited on Jun 10

Upload 5 files

7803d72
verified

CompressedGemma commited on May 21

Upload 3 files

766f12c
verified

CompressedGemma commited on May 14

Qwen attention tensors

44e6b86
verified

CompressedGemma commited on May 10

Update code comments

262cc7b
verified

CompressedGemma commited on May 7

Tensor tweak

dc3b370
verified

CompressedGemma commited on May 7

Experimental support for other LLMs

ae8c38d
verified

CompressedGemma commited on May 7

It's only calibrated for Gemma, atm.

07b428c
verified

CompressedGemma commited on May 6