Update app.py

app.py

@@ -446,6 +446,7 @@ model = AutoModel.from_pretrained(
     torch_dtype=torch.bfloat16,
     low_cpu_mem_usage=True,
     trust_remote_code=True,
+    use_flash_attn=False,
 ).eval().cuda()
 tokenizer = AutoTokenizer.from_pretrained("khang119966/Vintern-1B-v3_5-explainableAI", trust_remote_code=True, use_fast=False)
 
@@ -559,18 +560,11 @@ with gr.Blocks() as demo:
     gr.Markdown("""## 🎥 Visualizing How Multimodal Models Think
 This tool generates a video to **visualize how a multimodal model (image + text)** attends to different parts of an image while generating text.
 ### 📌 What it does:
-- Takes an input image and a text prompt.
-- Shows how the model’s attention shifts on the image for each generated token.
-- Helps explain the model’s behavior and decision-making.
-### 🖼️ Video layout (per frame):
-Each frame in the video includes:
-1. 🔥 **Heatmap over image**: Shows which area the model focuses on.
-2. 📝 **Generated text**: With old context, current token highlighted.
-3. 📊 **Token prediction table**: Shows the model’s top next-token guesses.
+- Takes an input image and a text prompt. - Shows how the model’s attention shifts on the image for each generated token. - Helps explain the model’s behavior and decision-making.
+### 🖼️ Video layout (per frame): 
+Each frame in the video includes: 1. 🔥 **Heatmap over image**: Shows which area the model focuses on. 2. 📝 **Generated text**: With old context, current token highlighted. 3. 📊 **Token prediction table**: Shows the model’s top next-token guesses.
 ### 🎯 Use cases:
-- Research explainability of vision-language models.
-- Debugging or interpreting model outputs.
-- Creating educational visualizations.
+- Research explainability of vision-language models. - Debugging or interpreting model outputs. - Creating educational visualizations.
 """)
     
     with gr.Row():