Update README.md

README.md

@@ -14,7 +14,7 @@ datasets:
 - NilanE/SmallParallelDocs-Ja_En-6k-preChunkedForTinyllama
 ---
 
-In-progess long-context Japanese-English translation model based on tinyllama. Input should be 500-1000 tokens long.
+In-progess long-context Japanese-English translation model based on tinyllama. Input should be 500-1000 tokens long. Make sure to set 'do_sample = False' if using HF transformers for inference, or otherwise set temperature to 0 for deterministic outputs.
 
 ## Prompt format
 """Translate this from Japanese to English:\n### JAPANESE: {source text} \n### ENGLISH: """