Update app.py

app.py

@@ -2,6 +2,8 @@ import spaces
 import gradio as gr
 from transformers import AutoTokenizer, LlamaForCausalLM
 import os
+import fastchat
+from fastchat.conversation import get_conv_template
 
 tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-2-7b-chat-hf", use_auth_token=os.getenv("HUGGINGFACE_TOKEN"))
 model = LlamaForCausalLM.from_pretrained("kaist-ai/Prometheus-13b-v1.0", device_map="auto", load_in_8bit=True)
@@ -14,14 +16,18 @@ An instruction (might include an Input inside it), a response to evaluate, a ref
 2. After writing a feedback, write a score that is an integer between 1 and 5. You should refer to the score rubric.
 3. The output format should look as follows: \"Feedback: (write a feedback for criteria) [RESULT] (an integer number between 1 and 5)\"
 4. Please do not generate any other opening, closing, and explanations.
+
 ###The instruction to evaluate:
 {instruction_to_evaluate}
+
 ###Response to evaluate:
 {response_to_evaluate} 
+
 ###Reference Answer (Score 5):
 {reference_answer} 
+
 ###Score Rubrics:
-[{criteria_description}]
+{criteria_description}
 Score 1: {score1_description}
 Score 2: {score2_description}
 Score 3: {score3_description}
@@ -31,7 +37,7 @@ Score 5: {score5_description}
 
     input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to("cuda")
 
-    outputs = model.generate(input_ids)
+    outputs = model.generate(input_ids, sample=True, temperature=1.0, top_p=0.9, max_new_tokens=650, repetition_penalty=1.03)
     result = tokenizer.decode(outputs[0])
 
     return result
@@ -49,7 +55,7 @@ iface = gr.Interface(
         gr.Textbox(label="Score 4 Description", placeholder="Enter Score 4 Description Here...", lines=2),
         gr.Textbox(label="Score 5 Description", placeholder="Enter Score 5 Description Here...", lines=2)
     ],
-    outputs="text",
+    outputs="🎏KAIST-AI/⚖️Prometheus",
     title="Welcome to🌟Tonic's⚖️Prometheus",
     description="[🎏KAIST-AI/⚖️Prometheus](https://huggingface.co/kaist-ai/prometheus-13b-v1.0) Prometheus is an alternative of GPT-4 evaluation when doing fine-grained evaluation of an underlying LLM & a Reward model for Reinforcement Learning from Human Feedback (RLHF). You can use this demo to try out their model !  You can also use [🎏KAIST-AI/⚖️Prometheus](https://huggingface.co/kaist-ai/prometheus-13b-v1.0) [by cloning this space](https://huggingface.co/spaces/Tonic/prometheus/tree/main?clone=true). [🧬🔬🔍 Simply click here: 🤗](https://huggingface.co/spaces/Tonic/prometheus?duplicate=true) Join us : 🌟TeamTonic🌟 is always making cool demos! Join our active builder's🛠️community 👻  [![Join us on Discord](https://img.shields.io/discord/1109943800132010065?label=Discord&logo=discord&style=flat-square)](https://discord.gg/GWpVpekp) On 🤗Huggingface: [TeamTonic](https://huggingface.co/TeamTonic) & [MultiTransformer](https://huggingface.co/MultiTransformer) On 🌐Github: [Tonic-AI](https://github.com/tonic-ai) & contribute to 🌟 [DataTonic](https://github.com/Tonic-AI/DataTonic) 🤗Big thanks to Yuvi Sharma and all the folks at huggingface for the community grant 🤗",
     examples=[