stabilityai
/

stablelm-2-12b-chat

Text Generation

Inference Endpoints

Model card Files Files and versions Community

pvduy commited on Apr 8

Commit

56b4bc6

•

1 Parent(s): 3f24747

Update README.md

Files changed (1) hide show

README.md +2 -1

README.md CHANGED Viewed

@@ -54,7 +54,8 @@ tokens = model.generate(
     inputs.to(model.device),
     max_new_tokens=100,
     temperature=0.7,
-    do_sample=True
 )
 output = tokenizer.decode(tokens[:, inputs.shape[-1]:][0], skip_special_tokens=False)

     inputs.to(model.device),
     max_new_tokens=100,
     temperature=0.7,
+    do_sample=True,
+    eos_token_id=100278 # <|im_end|>
 )
 output = tokenizer.decode(tokens[:, inputs.shape[-1]:][0], skip_special_tokens=False)