fix generation

app.py

@@ -1,3 +1,4 @@
+import os
 import torch
 import spaces
 from collections.abc import Iterator
@@ -9,6 +10,7 @@ MAX_MAX_NEW_TOKENS = 4096
 DEFAULT_MAX_NEW_TOKENS = 2048
 MAX_INPUT_TOKEN_LENGTH = 4096
 
+HF_TOKEN = os.environ['HF_TOKEN']
 
 DESCRIPTION = """\
 # IndicTrans3-beta 🚀
@@ -20,8 +22,8 @@ DESCRIPTION = """\
 
 # if torch.cuda.is_available():
 model_id = "ai4bharat/IndicTrans3-beta"
-model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float16, device_map="auto", offload_folder="offload")
-tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float16, device_map="auto", offload_folder="offload", token=HF_TOKEN)
+tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-3.2-3B-Instruct")
 
 
 LANGUAGES = {
@@ -180,7 +182,7 @@ with gr.Blocks(theme=gr.themes.Default(), css=css) as demo:
                 gr.Number(value=0.1, visible=False),
                 gr.Number(value=0.9, visible=False),
                 gr.Number(value=50, visible=False),
-                gr.Number(value=0.1, visible=False)
+                gr.Number(value=1.0, visible=False)
             ], 
             outputs=text_output
         )
@@ -197,10 +199,10 @@ with gr.Blocks(theme=gr.themes.Default(), css=css) as demo:
                 tgt_lang, 
                 text_input,
                 gr.Number(value=4096, visible=False),
-                gr.Number(value=0, visible=False),
+                gr.Number(value=0.1, visible=False),
                 gr.Number(value=0.9, visible=False),
                 gr.Number(value=50, visible=False),
-                gr.Number(value=0.0, visible=False)
+                gr.Number(value=1.0, visible=False)
             ], 
             outputs=text_output,
             fn=generate_for_examples,