Update app.py

app.py

@@ -5,19 +5,24 @@ def generate_text(prompt, max_length, temperature, top_p, repetition_penalty):
     tokenizer = GPT2TokenizerFast.from_pretrained("JakeTurner616/Adonalsium-gpt-neo-1.3B")
     model = GPTNeoForCausalLM.from_pretrained("JakeTurner616/Adonalsium-gpt-neo-1.3B")
 
-    # Check if tokenizer has a padding token, if not, set it to eos_token
+    # Set pad token
     if tokenizer.pad_token is None:
         tokenizer.pad_token = tokenizer.eos_token
-        model.resize_token_embeddings(len(tokenizer))
 
-    inputs = tokenizer(prompt, return_tensors="pt", padding=True)
+    inputs = tokenizer(prompt, return_tensors="pt", padding=True, truncation=True)
+    # Ensure that pad_token_id is set for the model
+    model.config.pad_token_id = tokenizer.pad_token_id
+
     outputs = model.generate(
         input_ids=inputs["input_ids"],
+        attention_mask=inputs["attention_mask"],  # Explicitly pass attention mask
         max_length=max_length,
         temperature=temperature,
         top_p=top_p,
         repetition_penalty=repetition_penalty,
-        no_repeat_ngram_size=2
+        no_repeat_ngram_size=2,  # Correctly specify the no_repeat_ngram_size
+        do_sample=True,  # Ensure sampling is enabled for temperature and top_p to take effect
+        pad_token_id=tokenizer.pad_token_id
     )
 
     generated_texts = [tokenizer.decode(output, skip_special_tokens=True) for output in outputs]