Update README.md
Browse files
README.md
CHANGED
|
@@ -60,7 +60,7 @@ messages = [
|
|
| 60 |
|
| 61 |
prompts = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)
|
| 62 |
|
| 63 |
-
llm = LLM(model=model_id, tensor_parallel_size=number_gpus)
|
| 64 |
|
| 65 |
outputs = llm.generate(prompts, sampling_params)
|
| 66 |
|
|
|
|
| 60 |
|
| 61 |
prompts = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)
|
| 62 |
|
| 63 |
+
llm = LLM(model=model_id, tensor_parallel_size=number_gpus, max_model_len=4096)
|
| 64 |
|
| 65 |
outputs = llm.generate(prompts, sampling_params)
|
| 66 |
|