change model

app.py

@@ -20,11 +20,14 @@ model = GPT2Model.from_pretrained('gpt2')
 # Generate text using the model and tokenizer
 #@spaces.GPU(duration=60)
 def generate_text(input_text):
-    input_ids = tokenizer.encode(input_text, return_tensors="pt")#.to("cuda")
+    #input_ids = tokenizer.encode(input_text, return_tensors="pt")#.to("cuda")
     #attention_mask = input_ids.ne(tokenizer.pad_token_id).long()
     #output = model.generate(input_ids, max_new_tokens=512, top_k=50, top_p=0.95, temperature=0.7, do_sample=True)# attention_mask=attention_mask, max_length=100, num_return_sequences=1, no_repeat_ngram_size=2, top_k=50, top_p=0.95, temperature=0.7, do_sample=True)
-    output = model.generate(input_ids) #, attention_mask=attention_mask, max_length=100, num_return_sequences=1, no_repeat_ngram_size=2, top_k=50, top_p=0.95, temperature=0.7, do_sample=True)
-    return tokenizer.decode(output[0])
+    #output = model.generate(input_ids) #, attention_mask=attention_mask, max_length=100, num_return_sequences=1, no_repeat_ngram_size=2, top_k=50, top_p=0.95, temperature=0.7, do_sample=True)
+    #return tokenizer.decode(output[0])
+    encoded_input = tokenizer(input_text, return_tensors='pt')
+    output = model(**encoded_input)
+    return output
 
 # Example of disabling Exllama backend (if applicable in your configuration)
 #config = {"disable_exllama": True}