Adding quantized models, quantization code and experiments, as well as composite benchmark data code

This commit is contained in:
emilyworks 2024-12-05 02:27:28 -05:00
parent 17c8f9b8dc
commit c8b043675a
2 changed files with 7050 additions and 0 deletions

View File

@ -0,0 +1,3 @@
This subdirectory holds the code, experiments, and data used to create and finetune various quantized llama3.2-3b models.
The actual models that resulted from the above experiments can be found [here](https://drive.google.com/drive/folders/101iMawJ5dXQu2YehHPyviDDM_Q3i6Np0?usp=sharing).

File diff suppressed because one or more lines are too long