Update app.py

app.py

@@ -14,22 +14,22 @@ 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=0.5, max_new_tokens=50):
+def generate_response(input_text, temperature=0.2, max_new_tokens=20):
     input_ids = tokenizer.encode(input_text, return_tensors='pt').to(model.device)
 
     if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH:
         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=120.0, skip_prompt=True, skip_special_tokens=True)
     generate_kwargs = dict(
         input_ids=input_ids,
         streamer=streamer,
         max_new_tokens=max_new_tokens,
         do_sample=True,
-        top_k=40,
+        top_k=20,
         top_p=0.9,
         temperature=temperature,
         eos_token_id=[tokenizer.eos_token_id]