kiseich
/

llm-jp-3-13b-Etask

Inference Endpoints

Model card Files Files and versions Community

kiseich commited on 4 days ago

Commit

1e15c2f

•

1 Parent(s): a626b75

fix number of tokens

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -87,7 +87,7 @@ for data in tqdm(datasets):
       outputs = model.generate(
           tokenized_input,
           attention_mask=attention_mask,
-          max_new_tokens=512,
           do_sample=False,
           repetition_penalty=1.2,
           pad_token_id=tokenizer.eos_token_id

       outputs = model.generate(
           tokenized_input,
           attention_mask=attention_mask,
+          max_new_tokens=100,
           do_sample=False,
           repetition_penalty=1.2,
           pad_token_id=tokenizer.eos_token_id