RedHatAI
/

granite-3.1-8b-base-quantized.w4a16

Text Generation

compressed-tensors

Model card Files Files and versions Community

nm-research commited on Jan 20

Commit

02e39c0

·

verified ·

1 Parent(s): 646956f

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -143,8 +143,8 @@ oneshot(
 )
 # Save to disk compressed.
-model.save_pretrained(SAVE_DIR, save_compressed=True)
-tokenizer.save_pretrained(SAVE_DIR)
 ```
 ## Evaluation

 )
 # Save to disk compressed.
+model.save_pretrained(quant_path, save_compressed=True)
+tokenizer.save_pretrained(quant_path)
 ```
 ## Evaluation