Update app.py

app.py

@@ -14,7 +14,7 @@ tokenizer = AutoTokenizer.from_pretrained(model_id)
 model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")
 tokenizer.pad_token = tokenizer.eos_token
 
-MAX_INPUT_TOKEN_LENGTH = 4096
+MAX_INPUT_TOKEN_LENGTH = 10000
 
 def generate_response(input_text, temperature=0.7, max_new_tokens=100):
     input_ids = tokenizer.encode(input_text, return_tensors='pt').to(model.device)
@@ -23,7 +23,7 @@ def generate_response(input_text, temperature=0.7, max_new_tokens=100):
         input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:]
         st.warning(f"Se recortó la entrada porque excedió el límite de {MAX_INPUT_TOKEN_LENGTH} tokens.")
 
-    streamer = TextIteratorStreamer(tokenizer, timeout=30.0, skip_prompt=True, skip_special_tokens=True)
+    streamer = TextIteratorStreamer(tokenizer, timeout=60.0, skip_prompt=True, skip_special_tokens=True)
     generate_kwargs = dict(
         input_ids=input_ids,
         streamer=streamer,
@@ -65,7 +65,7 @@ def main():
             st.write("Archivo CSV cargado exitosamente:")
             st.write(df.head())
 
-            initial_prompt = f"Here is a list of job titles: {job_titles}. Could you extract the first job title from this list and print it clearly?"
+            initial_prompt = f"The following is a list of job titles: {job_titles}. Please provide the first job title from this list."
             st.write(f"Query: {query}")
             st.write(f"Prompt inicial: {initial_prompt}")