Diksha2001 commited on
Commit
e44e2bf
1 Parent(s): 546c728

Update Finetuning_Pipeline.py

Browse files
Files changed (1) hide show
  1. Finetuning_Pipeline.py +2 -2
Finetuning_Pipeline.py CHANGED
@@ -62,8 +62,8 @@ def start_vllm_server(model_name):
62
  "vllm",
63
  "serve",
64
  "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
65
- "--gpu_memory_utilization=0.98",
66
- "--max_model_len=4096",
67
  "--enable-chunked-prefill=False",
68
  "--num_scheduler_steps=2"
69
  ]
 
62
  "vllm",
63
  "serve",
64
  "hugging-quants/Meta-Llama-3.1-8B-Instruct-AWQ-INT4",
65
+ "--gpu_memory_utilization=0.99",
66
+ "--max_model_len=8192",
67
  "--enable-chunked-prefill=False",
68
  "--num_scheduler_steps=2"
69
  ]