AWQ is an efficient, accurate and blazing-fast low-bit weight quantization method, currently supporting 4-bit quantization. Compared to GPTQ, it offers faster Transformers-based inference with equivalent or better quality compared to the most commonly used GPTQ settings.

It is supported by:

Text Generation Webui - using Loader: AutoAWQ
vLLM - Llama and Mistral models only
Hugging Face Text Generation Inference (TGI)
Transformers version 4.35.0 and later, from any code or client that supports Transformers
AutoAWQ - for use from Python code

Repositories available

Prompt template: Prometheus

###Task Description:
An instruction (might include an Input inside it), a response to evaluate, a reference answer that gets a score of 5, and a score rubric representing a evaluation criteria are given.
1. Write a detailed feedback that assess the quality of the response strictly based on the given score rubric, not evaluating in general.
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:
{prompt}

###Response to evaluate:
{{response}}

###Reference Answer (Score 5):
{{reference_answer}}

###Score Rubrics:
[{{criteria_description}}]
Score 1: {{score1_description}}
Score 2: {{score2_description}}
Score 3: {{score3_description}}
Score 4: {{score4_description}}
Score 5: {{score5_description}}

###Feedback:

Licensing

The creator of the source model has listed its license as apache-2.0, and this quantization has therefore used that same license.

As this model is based on Llama 2, it is also subject to the Meta Llama 2 license terms, and the license files for that are additionally included. It should therefore be considered as being claimed to be licensed under both licenses. I contacted Hugging Face for clarification on dual licensing but they do not yet have an official position. Should this change, or should Meta provide any feedback on this situation, I will update this section accordingly.

In the meantime, any questions regarding licensing, and in particular how these two licenses might interact, should be directed to the original model repository: KAIST AI's Prometheus 13B V1.0.

Provided files, and AWQ parameters

I currently release 128g GEMM models only. The addition of group_size 32 models, and GEMV kernel models, is being actively considered.

Models are released as sharded safetensors files.

Branch	Bits	GS	AWQ Dataset	Seq Len	Size
main	4	128	wikitext	4096	7.25 GB

How to easily download and use this model in text-generation-webui

Please make sure you're using the latest version of text-generation-webui.

It is strongly recommended to use the text-generation-webui one-click-installers unless you're sure you know how to make a manual install.

Click the Model tab.
Under Download custom model or LoRA, enter TheBloke/prometheus-13B-v1.0-AWQ.
Click Download.
The model will start downloading. Once it's finished it will say "Done".
In the top left, click the refresh icon next to Model.
In the Model dropdown, choose the model you just downloaded: prometheus-13B-v1.0-AWQ
Select Loader: AutoAWQ.
Click Load, and the model will load and is now ready for use.
If you want any custom settings, set them and then click Save settings for this model followed by Reload the Model in the top right.
Once you're ready, click the Text Generation tab and enter a prompt to get started!

Multi-user inference server: vLLM

Documentation on installing and using vLLM can be found here.

Please ensure you are using vLLM version 0.2 or later.
When using vLLM as a server, pass the --quantization awq parameter.

For example:

python3 -m vllm.entrypoints.api_server --model TheBloke/prometheus-13B-v1.0-AWQ --quantization awq --dtype auto

When using vLLM from Python code, again set quantization=awq.

For example:

python

from vllm import LLM, SamplingParams

prompts = [
    "Tell me about AI",
    "Write a story about llamas",
    "What is 291 - 150?",
    "How much wood would a woodchuck chuck if a woodchuck could chuck wood?",
]
prompt_template=f'''###Task Description:
An instruction (might include an Input inside it), a response to evaluate, a reference answer that gets a score of 5, and a score rubric representing a evaluation criteria are given.
1. Write a detailed feedback that assess the quality of the response strictly based on the given score rubric, not evaluating in general.
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:
{prompt}

###Response to evaluate:
{{response}}

###Reference Answer (Score 5):
{{reference_answer}}

###Score Rubrics:
[{{criteria_description}}]
Score 1: {{score1_description}}
Score 2: {{score2_description}}
Score 3: {{score3_description}}
Score 4: {{score4_description}}
Score 5: {{score5_description}}

###Feedback:
'''

prompts = [prompt_template.format(prompt=prompt) for prompt in prompts]

sampling_params = SamplingParams(temperature=0.8, top_p=0.95)

llm = LLM(model="TheBloke/prometheus-13B-v1.0-AWQ", quantization="awq", dtype="auto")

outputs = llm.generate(prompts, sampling_params)

# Print the outputs.
for output in outputs:
    prompt = output.prompt
    generated_text = output.outputs[0].text
    print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")

Multi-user inference server: Hugging Face Text Generation Inference (TGI)

Use TGI version 1.1.0 or later. The official Docker container is: ghcr.io/huggingface/text-generation-inference:1.1.0

Example Docker parameters:

--model-id TheBloke/prometheus-13B-v1.0-AWQ --port 3000 --quantize awq --max-input-length 3696 --max-total-tokens 4096 --max-batch-prefill-tokens 4096

Example Python code for interfacing with TGI (requires huggingface-hub 0.17.0 or later):

pip3 install huggingface-hub

python

from huggingface_hub import InferenceClient

endpoint_url = "https://your-endpoint-url-here"

prompt = "Tell me about AI"
prompt_template=f'''###Task Description:
An instruction (might include an Input inside it), a response to evaluate, a reference answer that gets a score of 5, and a score rubric representing a evaluation criteria are given.
1. Write a detailed feedback that assess the quality of the response strictly based on the given score rubric, not evaluating in general.
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:
{prompt}

###Response to evaluate:
{{response}}

###Reference Answer (Score 5):
{{reference_answer}}

###Score Rubrics:
[{{criteria_description}}]
Score 1: {{score1_description}}
Score 2: {{score2_description}}
Score 3: {{score3_description}}
Score 4: {{score4_description}}
Score 5: {{score5_description}}

###Feedback:
'''

client = InferenceClient(endpoint_url)
response = client.text_generation(prompt,
                                  max_new_tokens=128,
                                  do_sample=True,
                                  temperature=0.7,
                                  top_p=0.95,
                                  top_k=40,
                                  repetition_penalty=1.1)

print(f"Model output: ", response)

Inference from Python code using Transformers

Install the necessary packages

Requires: Transformers 4.35.0 or later.
Requires: AutoAWQ 0.1.6 or later.

pip3 install --upgrade "autoawq>=0.1.6" "transformers>=4.35.0"

Note that if you are using PyTorch 2.0.1, the above AutoAWQ command will automatically upgrade you to PyTorch 2.1.0.

If you are using CUDA 11.8 and wish to continue using PyTorch 2.0.1, instead run this command:

pip3 install https://github.com/casper-hansen/AutoAWQ/releases/download/v0.1.6/autoawq-0.1.6+cu118-cp310-cp310-linux_x86_64.whl

If you have problems installing AutoAWQ using the pre-built wheels, install it from source instead:

shell

pip3 uninstall -y autoawq
git clone https://github.com/casper-hansen/AutoAWQ
cd AutoAWQ
pip3 install .

Transformers example code (requires Transformers 4.35.0 and later)

python

from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer

model_name_or_path = "TheBloke/prometheus-13B-v1.0-AWQ"

tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
model = AutoModelForCausalLM.from_pretrained(
    model_name_or_path,
    low_cpu_mem_usage=True,
    device_map="cuda:0"
)

# Using the text streamer to stream output one token at a time
streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)

prompt = "Tell me about AI"
prompt_template=f'''###Task Description:
An instruction (might include an Input inside it), a response to evaluate, a reference answer that gets a score of 5, and a score rubric representing a evaluation criteria are given.
1. Write a detailed feedback that assess the quality of the response strictly based on the given score rubric, not evaluating in general.
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:
{prompt}

###Response to evaluate:
{{response}}

###Reference Answer (Score 5):
{{reference_answer}}

###Score Rubrics:
[{{criteria_description}}]
Score 1: {{score1_description}}
Score 2: {{score2_description}}
Score 3: {{score3_description}}
Score 4: {{score4_description}}
Score 5: {{score5_description}}

###Feedback:
'''

# Convert prompt to tokens
tokens = tokenizer(
    prompt_template,
    return_tensors='pt'
).input_ids.cuda()

generation_params = {
    "do_sample": True,
    "temperature": 0.7,
    "top_p": 0.95,
    "top_k": 40,
    "max_new_tokens": 512,
    "repetition_penalty": 1.1
}

# Generate streamed output, visible one token at a time
generation_output = model.generate(
    tokens,
    streamer=streamer,
    **generation_params
)

# Generation without a streamer, which will include the prompt in the output
generation_output = model.generate(
    tokens,
    **generation_params
)

# Get the tokens from the output, decode them, print them
token_output = generation_output[0]
text_output = tokenizer.decode(token_output)
print("model.generate output: ", text_output)

# Inference is also possible via Transformers' pipeline
from transformers import pipeline

pipe = pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    **generation_params
)

pipe_output = pipe(prompt_template)[0]['generated_text']
print("pipeline output: ", pipe_output)

Compatibility

The files provided are tested to work with:

text-generation-webui using Loader: AutoAWQ.
vLLM version 0.2.0 and later.
Hugging Face Text Generation Inference (TGI) version 1.1.0 and later.
Transformers version 4.35.0 and later.
AutoAWQ version 0.1.1 and later.

Discord

For further support, and discussions on these models and AI in general, join us at:

TheBloke AI's Discord server

Thanks, and how to contribute

Thanks to the chirper.ai team!

Thanks to Clay from gpus.llm-utils.org!

I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.

If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.

Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.

Patreon: https://patreon.com/TheBlokeAI
Ko-Fi: https://ko-fi.com/TheBlokeAI

Special thanks to: Aemon Algiz.

Patreon special mentions: Brandon Frisco, LangChain4j, Spiking Neurons AB, transmissions 11, Joseph William Delisle, Nitin Borwankar, Willem Michiel, Michael Dempsey, vamX, Jeffrey Morgan, zynix, jjj, Omer Bin Jawed, Sean Connelly, jinyuan sun, Jeromy Smith, Shadi, Pawan Osman, Chadd, Elijah Stavena, Illia Dulskyi, Sebastain Graf, Stephen Murray, terasurfer, Edmond Seymore, Celu Ramasamy, Mandus, Alex, biorpg, Ajan Kanaga, Clay Pascal, Raven Klaugh, 阿明, K, ya boyyy, usrbinkat, Alicia Loh, John Villwock, ReadyPlayerEmma, Chris Smitley, Cap'n Zoog, fincy, GodLy, S_X, sidney chen, Cory Kujawski, OG, Mano Prime, AzureBlack, Pieter, Kalila, Spencer Kim, Tom X Nguyen, Stanislav Ovsiannikov, Michael Levine, Andrey, Trailburnt, Vadim, Enrico Ros, Talal Aujan, Brandon Phillips, Jack West, Eugene Pentland, Michael Davis, Will Dee, webtim, Jonathan Leane, Alps Aficionado, Rooh Singh, Tiffany J. Kim, theTransient, Luke @flexchar, Elle, Caitlyn Gatomon, Ari Malik, subjectnull, Johann-Peter Hartmann, Trenton Dambrowitz, Imad Khwaja, Asp the Wyvern, Emad Mostaque, Rainer Wilmers, Alexandros Triantafyllidis, Nicholas, Pedro Madruga, SuperWojo, Harry Royden McLaughlin, James Bentley, Olakabola, David Ziegler, Ai Maven, Jeff Scroggin, Nikolai Manek, Deo Leter, Matthew Berman, Fen Risland, Ken Nordquist, Manuel Alberto Morcote, Luke Pendergrass, TL, Fred von Graf, Randy H, Dan Guido, NimbleBox.ai, Vitor Caleffi, Gabriel Tamborski, knownsqashed, Lone Striker, Erik Bjäreholt, John Detwiler, Leonard Tan, Iucharbius

Thank you to all my generous patrons and donaters!

And thank you again to a16z for their generous grant.

Original model card: KAIST AI's Prometheus 13B V1.0

Links for Reference

Homepage:https://github.com/kaistAI/Prometheus
Repository:https://github.com/kaistAI/Prometheus
Paper:https://arxiv.org/abs/2310.08491
Point of Contact:seungone@kaist.ac.kr

TL;DR

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).

Prometheus is a language model using Llama-2-Chat as a base model and fine-tuned on 100K feedback within the Feedback Collection. Since it was fine-tuned on a large amount of feedback, it is specialized at evaluating long-form responses, outperforming GPT-3.5-Turbo, Llama-2-Chat 70B, and on par with GPT-4 on various benchmarks. Most importantly, this was possible since we appended 2 reference materials (reference answer, and customized score rubric). Prometheus is a cheap and powerful alternative to GPT-4 evaluation, which one could use to evaluate LLMs with customized criteria (e.g., Child readability, Cultural Sensitivity, Creativity). Also, it could be used as a reward model for Reinforcement Learning from Human Feedback (RLHF).

Model Details

Model Description

Model type: Language model
Language(s) (NLP): English
License: Apache 2.0
Related Models: All Prometheus Checkpoints
Resources for more information:
- Research paper
- GitHub Repo

Prometheus is trained with two different sizes (7B and 13B). You could check the 7B sized LM on this page. Also, check out our dataset as well on this page.

Prompt Format

Prometheus requires 4 components in the input: An instruction, a response to evaluate, a score rubric, and a reference answer. You could refer to the prompt format below. You should fill in the instruction, response, reference answer, criteria description, and score description for score in range of 1 to 5.

###Task Description:
An instruction (might include an Input inside it), a response to evaluate, a reference answer that gets a score of 5, and a score rubric representing a evaluation criteria are given.
1. Write a detailed feedback that assess the quality of the response strictly based on the given score rubric, not evaluating in general.
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}

###Response to evaluate:
{response}

###Reference Answer (Score 5):
{reference_answer}

###Score Rubrics:
[{criteria_description}]
Score 1: {score1_description}
Score 2: {score2_description}
Score 3: {score3_description}
Score 4: {score4_description}
Score 5: {score5_description}

###Feedback:

After this, you should apply the conversation template of Llama-2-Chat (not applying it might lead to unexpected behaviors). You can find the conversation class at this link.

conv = get_conv_template("llama-2")
conv.set_system_message("You are a fair evaluator language model.")
conv.append_message(conv.roles[0], dialogs['instruction'])
conv.append_message(conv.roles[1], None)
prompt = conv.get_prompt()

x = tokenizer(prompt,truncation=False)

As a result, a feedback and score decision will be generated, divided by a separating phrase [RESULT]

License

Feedback Collection and Prometheus is subject to OpenAI's Terms of Use for the generated data. If you suspect any violations, please reach out to us.

Usage

Find below some example scripts on how to use the model in transformers:

Using the Pytorch model

Running the model on a CPU

Click to expand

python

from transformers import T5Tokenizer, T5ForConditionalGeneration

tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-2-7b-chat-hf")
model = LlamaForCausalLM.from_pretrained("kaist-ai/Prometheus-13b-v1.0", device_map="auto")

input_text = "###Task Description: An instruction (might include an Input inside it), a response to evaluate, a reference answer that gets a score of 5, and a score rubric representing a evaluation criteria are given. 1. Write a detailed feedback that assess the quality of the response strictly based on the given score rubric, not evaluating in general. 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: You are a well-known psychiatrist who has a reputation for being empathetic and understanding. A client comes to you saying they've had a really hard day at work. They describe their boss as being overly critical and not listening to their ideas, which has left them feeling frustrated and undervalued. They also mention feeling overwhelmed with the workload and a growing sense of loneliness as they are new to the city. How do you respond to them to make them feel heard and understood, as well as offering helpful advice? ###Response to evaluate: It's indeed challenging to deal with a difficult boss and to carry a large workload, especially when you are new to a city and haven't established a supportive network. I would suggest taking a step back to think about your situation and perhaps speak to your boss regarding the issues you're facing. On the other hand, dealing with a large workload can be managed by prioritizing tasks, and maybe you can discuss it with your superiors too. In regards to feeling lonely, you might want to explore activities or groups that align with your interests to meet new people. And also, remember to take time for yourself to relax and destress. I'm here to assist you as we navigate through these issues. ###Reference Answer (Score 5): I hear that you're dealing with quite a few challenging situations at once. It must be tough to have your ideas dismissed by your boss and to feel undervalued in your work environment. It's understandable that such experiences can lead to frustration and a feeling of being overwhelmed. It's important to acknowledge these feelings and not to dismiss them. Feeling heard and validated in our experiences can often be the first step towards resolving the challenges we face. So, I want you to know that your feelings are completely valid and it's okay to be frustrated and overwhelmed. In terms of dealing with your boss, have you considered seeking a private conversation with them to express your feelings? It's often beneficial to communicate what you're experiencing, using "I" statements to express how you feel when your ideas are not taken into consideration. This can lead to mutual understanding and possibly a change in behavior. About the workload, it might help to prioritize tasks and potentially delegate, if possible. Also, consider discussing your workload with your superiors. There might be a misunderstanding about what's manageable or they might have suggestions about how to handle the situation. On the personal front, feeling lonely, especially when you're new to a city, can be really hard. Seek out opportunities to meet new people, perhaps through hobbies, community activities, or online groups. It might take a bit of time, but gradually, you can build a network of friends and acquaintances. Remember, it's perfectly okay to have bad days and it's important to take care of your mental health. Consider incorporating activities into your daily routine that make you happy and help you unwind. This could be anything from reading, yoga, going for a walk, or even listening to your favorite music. Please know that you're not alone in this. I'm here to support you through this challenging time and together, we can work towards resolving these issues. ###Score Rubrics: [Is the model able to identify and react correctly to the emotional context of the user's input?] Score 1: The model utterly fails to grasp the user's emotional context and responds in an unfitting manner. Score 2: The model sporadically identifies the emotional context but frequently replies in a manner that doesn't match the user's emotional status. Score 3: The model typically identifies the emotional context and reacts suitably, but occasionally misreads or misjudges the user's feelings. Score 4: The model often identifies the emotional context and reacts suitably, with minor cases of misreading or misjudging. Score 5: The model flawlessly identifies the emotional context of the user's input and consistently responds in a considerate and empathetic manner. ###Feedback:"
input_ids = tokenizer(input_text, return_tensors="pt").input_ids

outputs = model.generate(input_ids)
print(tokenizer.decode(outputs[0]))

Running the model on a GPU

Click to expand

python

# pip install accelerate
import torch
from transformers import AutoTokenizer, LlamaForCausalLM

tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-2-7b-chat-hf")
model = LlamaForCausalLM.from_pretrained("kaist-ai/Prometheus-13b-v1.0", device_map="auto")

input_text = "###Task Description: An instruction (might include an Input inside it), a response to evaluate, a reference answer that gets a score of 5, and a score rubric representing a evaluation criteria are given. 1. Write a detailed feedback that assess the quality of the response strictly based on the given score rubric, not evaluating in general. 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: You are a well-known psychiatrist who has a reputation for being empathetic and understanding. A client comes to you saying they've had a really hard day at work. They describe their boss as being overly critical and not listening to their ideas, which has left them feeling frustrated and undervalued. They also mention feeling overwhelmed with the workload and a growing sense of loneliness as they are new to the city. How do you respond to them to make them feel heard and understood, as well as offering helpful advice? ###Response to evaluate: It's indeed challenging to deal with a difficult boss and to carry a large workload, especially when you are new to a city and haven't established a supportive network. I would suggest taking a step back to think about your situation and perhaps speak to your boss regarding the issues you're facing. On the other hand, dealing with a large workload can be managed by prioritizing tasks, and maybe you can discuss it with your superiors too. In regards to feeling lonely, you might want to explore activities or groups that align with your interests to meet new people. And also, remember to take time for yourself to relax and destress. I'm here to assist you as we navigate through these issues. ###Reference Answer (Score 5): I hear that you're dealing with quite a few challenging situations at once. It must be tough to have your ideas dismissed by your boss and to feel undervalued in your work environment. It's understandable that such experiences can lead to frustration and a feeling of being overwhelmed. It's important to acknowledge these feelings and not to dismiss them. Feeling heard and validated in our experiences can often be the first step towards resolving the challenges we face. So, I want you to know that your feelings are completely valid and it's okay to be frustrated and overwhelmed. In terms of dealing with your boss, have you considered seeking a private conversation with them to express your feelings? It's often beneficial to communicate what you're experiencing, using "I" statements to express how you feel when your ideas are not taken into consideration. This can lead to mutual understanding and possibly a change in behavior. About the workload, it might help to prioritize tasks and potentially delegate, if possible. Also, consider discussing your workload with your superiors. There might be a misunderstanding about what's manageable or they might have suggestions about how to handle the situation. On the personal front, feeling lonely, especially when you're new to a city, can be really hard. Seek out opportunities to meet new people, perhaps through hobbies, community activities, or online groups. It might take a bit of time, but gradually, you can build a network of friends and acquaintances. Remember, it's perfectly okay to have bad days and it's important to take care of your mental health. Consider incorporating activities into your daily routine that make you happy and help you unwind. This could be anything from reading, yoga, going for a walk, or even listening to your favorite music. Please know that you're not alone in this. I'm here to support you through this challenging time and together, we can work towards resolving these issues. ###Score Rubrics: [Is the model able to identify and react correctly to the emotional context of the user's input?] Score 1: The model utterly fails to grasp the user's emotional context and responds in an unfitting manner. Score 2: The model sporadically identifies the emotional context but frequently replies in a manner that doesn't match the user's emotional status. Score 3: The model typically identifies the emotional context and reacts suitably, but occasionally misreads or misjudges the user's feelings. Score 4: The model often identifies the emotional context and reacts suitably, with minor cases of misreading or misjudging. Score 5: The model flawlessly identifies the emotional context of the user's input and consistently responds in a considerate and empathetic manner. ###Feedback:"
input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to("cuda")

outputs = model.generate(input_ids, sample=True, temperature=1.0, top_p=0.9, max_new_tokens=256, repetition_penalty=1.03)
print(tokenizer.decode(outputs[0]))

Running the model on a GPU using different precisions

FP16

Click to expand

python

# pip install accelerate
import torch
from transformers import AutoTokenizer, LlamaForCausalLM

tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-2-7b-chat-hf")
model = LlamaForCausalLM.from_pretrained("kaist-ai/Prometheus-13b-v1.0", device_map="auto", torch_dtype=torch.float16)

input_text = "###Task Description: An instruction (might include an Input inside it), a response to evaluate, a reference answer that gets a score of 5, and a score rubric representing a evaluation criteria are given. 1. Write a detailed feedback that assess the quality of the response strictly based on the given score rubric, not evaluating in general. 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: You are a well-known psychiatrist who has a reputation for being empathetic and understanding. A client comes to you saying they've had a really hard day at work. They describe their boss as being overly critical and not listening to their ideas, which has left them feeling frustrated and undervalued. They also mention feeling overwhelmed with the workload and a growing sense of loneliness as they are new to the city. How do you respond to them to make them feel heard and understood, as well as offering helpful advice? ###Response to evaluate: It's indeed challenging to deal with a difficult boss and to carry a large workload, especially when you are new to a city and haven't established a supportive network. I would suggest taking a step back to think about your situation and perhaps speak to your boss regarding the issues you're facing. On the other hand, dealing with a large workload can be managed by prioritizing tasks, and maybe you can discuss it with your superiors too. In regards to feeling lonely, you might want to explore activities or groups that align with your interests to meet new people. And also, remember to take time for yourself to relax and destress. I'm here to assist you as we navigate through these issues. ###Reference Answer (Score 5): I hear that you're dealing with quite a few challenging situations at once. It must be tough to have your ideas dismissed by your boss and to feel undervalued in your work environment. It's understandable that such experiences can lead to frustration and a feeling of being overwhelmed. It's important to acknowledge these feelings and not to dismiss them. Feeling heard and validated in our experiences can often be the first step towards resolving the challenges we face. So, I want you to know that your feelings are completely valid and it's okay to be frustrated and overwhelmed. In terms of dealing with your boss, have you considered seeking a private conversation with them to express your feelings? It's often beneficial to communicate what you're experiencing, using "I" statements to express how you feel when your ideas are not taken into consideration. This can lead to mutual understanding and possibly a change in behavior. About the workload, it might help to prioritize tasks and potentially delegate, if possible. Also, consider discussing your workload with your superiors. There might be a misunderstanding about what's manageable or they might have suggestions about how to handle the situation. On the personal front, feeling lonely, especially when you're new to a city, can be really hard. Seek out opportunities to meet new people, perhaps through hobbies, community activities, or online groups. It might take a bit of time, but gradually, you can build a network of friends and acquaintances. Remember, it's perfectly okay to have bad days and it's important to take care of your mental health. Consider incorporating activities into your daily routine that make you happy and help you unwind. This could be anything from reading, yoga, going for a walk, or even listening to your favorite music. Please know that you're not alone in this. I'm here to support you through this challenging time and together, we can work towards resolving these issues. ###Score Rubrics: [Is the model able to identify and react correctly to the emotional context of the user's input?] Score 1: The model utterly fails to grasp the user's emotional context and responds in an unfitting manner. Score 2: The model sporadically identifies the emotional context but frequently replies in a manner that doesn't match the user's emotional status. Score 3: The model typically identifies the emotional context and reacts suitably, but occasionally misreads or misjudges the user's feelings. Score 4: The model often identifies the emotional context and reacts suitably, with minor cases of misreading or misjudging. Score 5: The model flawlessly identifies the emotional context of the user's input and consistently responds in a considerate and empathetic manner. ###Feedback:"
input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to("cuda")

outputs = model.generate(input_ids)
print(tokenizer.decode(outputs[0]))

INT8

Click to expand

python

# pip install bitsandbytes accelerate
from transformers import AutoTokenizer, LlamaForCausalLM

tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-2-7b-chat-hf")
model = LlamaForCausalLM.from_pretrained("kaist-ai/Prometheus-13b-v1.0", device_map="auto", load_in_8bit=True)

input_text = "###Task Description: An instruction (might include an Input inside it), a response to evaluate, a reference answer that gets a score of 5, and a score rubric representing a evaluation criteria are given. 1. Write a detailed feedback that assess the quality of the response strictly based on the given score rubric, not evaluating in general. 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: You are a well-known psychiatrist who has a reputation for being empathetic and understanding. A client comes to you saying they've had a really hard day at work. They describe their boss as being overly critical and not listening to their ideas, which has left them feeling frustrated and undervalued. They also mention feeling overwhelmed with the workload and a growing sense of loneliness as they are new to the city. How do you respond to them to make them feel heard and understood, as well as offering helpful advice? ###Response to evaluate: It's indeed challenging to deal with a difficult boss and to carry a large workload, especially when you are new to a city and haven't established a supportive network. I would suggest taking a step back to think about your situation and perhaps speak to your boss regarding the issues you're facing. On the other hand, dealing with a large workload can be managed by prioritizing tasks, and maybe you can discuss it with your superiors too. In regards to feeling lonely, you might want to explore activities or groups that align with your interests to meet new people. And also, remember to take time for yourself to relax and destress. I'm here to assist you as we navigate through these issues. ###Reference Answer (Score 5): I hear that you're dealing with quite a few challenging situations at once. It must be tough to have your ideas dismissed by your boss and to feel undervalued in your work environment. It's understandable that such experiences can lead to frustration and a feeling of being overwhelmed. It's important to acknowledge these feelings and not to dismiss them. Feeling heard and validated in our experiences can often be the first step towards resolving the challenges we face. So, I want you to know that your feelings are completely valid and it's okay to be frustrated and overwhelmed. In terms of dealing with your boss, have you considered seeking a private conversation with them to express your feelings? It's often beneficial to communicate what you're experiencing, using "I" statements to express how you feel when your ideas are not taken into consideration. This can lead to mutual understanding and possibly a change in behavior. About the workload, it might help to prioritize tasks and potentially delegate, if possible. Also, consider discussing your workload with your superiors. There might be a misunderstanding about what's manageable or they might have suggestions about how to handle the situation. On the personal front, feeling lonely, especially when you're new to a city, can be really hard. Seek out opportunities to meet new people, perhaps through hobbies, community activities, or online groups. It might take a bit of time, but gradually, you can build a network of friends and acquaintances. Remember, it's perfectly okay to have bad days and it's important to take care of your mental health. Consider incorporating activities into your daily routine that make you happy and help you unwind. This could be anything from reading, yoga, going for a walk, or even listening to your favorite music. Please know that you're not alone in this. I'm here to support you through this challenging time and together, we can work towards resolving these issues. ###Score Rubrics: [Is the model able to identify and react correctly to the emotional context of the user's input?] Score 1: The model utterly fails to grasp the user's emotional context and responds in an unfitting manner. Score 2: The model sporadically identifies the emotional context but frequently replies in a manner that doesn't match the user's emotional status. Score 3: The model typically identifies the emotional context and reacts suitably, but occasionally misreads or misjudges the user's feelings. Score 4: The model often identifies the emotional context and reacts suitably, with minor cases of misreading or misjudging. Score 5: The model flawlessly identifies the emotional context of the user's input and consistently responds in a considerate and empathetic manner. ###Feedback:"
input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to("cuda")

outputs = model.generate(input_ids)
print(tokenizer.decode(outputs[0]))

Citation

If you find the following model helpful, please consider citing our paper!

BibTeX:

bibtex

@misc{kim2023prometheus,
    title={Prometheus: Inducing Fine-grained Evaluation Capability in Language Models},
    author={Seungone Kim and Jamin Shin and Yejin Cho and Joel Jang and Shayne Longpre and Hwaran Lee and Sangdoo Yun and Seongjin Shin and Sungdong Kim and James Thorne and Minjoon Seo},
    year={2023},
    eprint={2310.08491},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}