Update app.py

app.py

@@ -20,7 +20,7 @@ def init_args():
     args.config_path = './config/llama_7b.json'
     args.spm_model_path = './model_file/tokenizer.model'
     args.batch_size = 1
-    args.seq_length = 512
+    args.seq_length = 256
     args.world_size = 1
     args.use_int8 = True
     args.top_p = 0
@@ -42,9 +42,6 @@ def init_model():
     model = load_model(model, args.load_model_path)
     model.eval()
 
-    # for name, parameter in model.named_parameters():
-    #     print(name)
-    #     print(parameter)
     print(torch.cuda.max_memory_allocated() / 1024 ** 3)
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     model.to(device)