RedHatAI
/

granite-3.1-2b-instruct-quantized.w4a16

Text Generation

compressed-tensors

Model card Files Files and versions

nm-research commited on Jan 20

Commit

7fd5827

·

verified ·

1 Parent(s): 1e0ff6c

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -142,8 +142,8 @@ 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