metadata

language:
  - en
license: cc-by-nc-4.0
library_name: transformers
tags:
  - reward model
  - RLHF
  - RLAIF
datasets:
  - berkeley-nest/Nectar
model-index:
  - name: Starling-LM-11B-alpha
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: AI2 Reasoning Challenge (25-Shot)
          type: ai2_arc
          config: ARC-Challenge
          split: test
          args:
            num_few_shot: 25
        metrics:
          - type: acc_norm
            value: 61.26
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CallComply/Starling-LM-11B-alpha
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: HellaSwag (10-Shot)
          type: hellaswag
          split: validation
          args:
            num_few_shot: 10
        metrics:
          - type: acc_norm
            value: 81.99
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CallComply/Starling-LM-11B-alpha
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU (5-Shot)
          type: cais/mmlu
          config: all
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 61.5
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CallComply/Starling-LM-11B-alpha
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: TruthfulQA (0-shot)
          type: truthful_qa
          config: multiple_choice
          split: validation
          args:
            num_few_shot: 0
        metrics:
          - type: mc2
            value: 41.53
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CallComply/Starling-LM-11B-alpha
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: Winogrande (5-shot)
          type: winogrande
          config: winogrande_xl
          split: validation
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 78.06
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CallComply/Starling-LM-11B-alpha
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GSM8k (5-shot)
          type: gsm8k
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 35.18
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CallComply/Starling-LM-11B-alpha
          name: Open LLM Leaderboard

Starling-LM-7B-alpha

Developed by: Banghua Zhu * , Evan Frick * , Tianhao Wu * , Hanlin Zhu and Jiantao Jiao.
Model type: Language Model finetuned with RLHF / RLAIF
License: Non commercial license
Finetuned from model: Openchat 3.5 (based on Mistral-7B-v0.1)

We introduce Starling-7B, an open large language model (LLM) trained by Reinforcement Learning from AI Feedback (RLAIF). The model harnesses the power of our new GPT-4 labeled ranking dataset, berkeley-nest/Nectar, and our new reward training and policy tuning pipeline. Starling-7B-alpha scores 8.09 in MT Bench with GPT-4 as a judge, outperforming every model to date on MT-Bench except for OpenAI's GPT-4 and GPT-4 Turbo. We release the ranking dataset Nectar, the reward model Starling-RM-7B-alpha and the language model Starling-LM-7B-alpha on HuggingFace, and an online demo in LMSYS Chatbot Arena. Stay tuned for our forthcoming code and paper, which will provide more details on the whole process.

Starling-LM-7B-alpha is a language model trained from Openchat 3.5 with reward model berkeley-nest/Starling-RM-7B-alpha and policy optimization method advantage-induced policy alignment (APA). The evaluation results are listed below.

Model	Tuning Method	MT Bench	AlpacaEval	MMLU
GPT-4-Turbo	?	9.32	97.70
GPT-4	SFT + PPO	8.99	95.28	86.4
Starling-7B	C-RLFT + APA	8.09	91.99	63.9
Claude-2	?	8.06	91.36	78.5
GPT-3.5-Turbo	?	7.94	89.37	70
Claude-1	?	7.9	88.39	77
Tulu-2-dpo-70b	SFT + DPO	7.89	95.1
Openchat-3.5	C-RLFT	7.81	88.51	64.3
Zephyr-7B-beta	SFT + DPO	7.34	90.60	61.4
Llama-2-70b-chat-hf	SFT + PPO	6.86	92.66	63
Neural-chat-7b-v3-1	SFT + DPO	6.84	84.53	62.4
Tulu-2-dpo-7b	SFT + DPO	6.29	85.1

For more detailed discussions, please check out our blog post, and stay tuned for our upcoming code and paper!

Blog: https://starling.cs.berkeley.edu/
Paper: Coming soon!
Code: Coming soon!

Uses

Important: Please use the exact chat template provided below for the model. Otherwise there will be a degrade in the performance. The model output can be verbose in rare cases. Please consider setting temperature = 0 to make this happen less.

Our model follows the exact chat template and usage as Openchat 3.5. Please refer to their model card for more details. In addition, our model is hosted on LMSYS Chatbot Arena for free test.

The conversation template is the same as Openchat 3.5:

import transformers
tokenizer = transformers.AutoTokenizer.from_pretrained("openchat/openchat_3.5")

# Single-turn
tokens = tokenizer("GPT4 Correct User: Hello<|end_of_turn|>GPT4 Correct Assistant:").input_ids
assert tokens == [1, 420, 6316, 28781, 3198, 3123, 1247, 28747, 22557, 32000, 420, 6316, 28781, 3198, 3123, 21631, 28747]

# Multi-turn
tokens = tokenizer("GPT4 Correct User: Hello<|end_of_turn|>GPT4 Correct Assistant: Hi<|end_of_turn|>GPT4 Correct User: How are you today?<|end_of_turn|>GPT4 Correct Assistant:").input_ids
assert tokens == [1, 420, 6316, 28781, 3198, 3123, 1247, 28747, 22557, 32000, 420, 6316, 28781, 3198, 3123, 21631, 28747, 15359, 32000, 420, 6316, 28781, 3198, 3123, 1247, 28747, 1602, 460, 368, 3154, 28804, 32000, 420, 6316, 28781, 3198, 3123, 21631, 28747]

# Coding Mode
tokens = tokenizer("Code User: Implement quicksort using C++<|end_of_turn|>Code Assistant:").input_ids
assert tokens == [1, 7596, 1247, 28747, 26256, 2936, 7653, 1413, 334, 1680, 32000, 7596, 21631, 28747]

Code Examples

import transformers

tokenizer = transformers.AutoTokenizer.from_pretrained("berkeley-nest/Starling-LM-7B-alpha")
model = transformers.AutoModelForCausalLM.from_pretrained("berkeley-nest/Starling-LM-7B-alpha")

def generate_response(prompt):
    input_ids = tokenizer(prompt, return_tensors="pt").input_ids
    outputs = model.generate(
        input_ids,
        max_length=256,
        pad_token_id=tokenizer.pad_token_id,
        eos_token_id=tokenizer.eos_token_id,
    )
    response_ids = outputs[0]
    response_text = tokenizer.decode(response_ids, skip_special_tokens=True)
    return response_text

# Single-turn conversation
prompt = "Hello, how are you?"
single_turn_prompt = f"GPT4 Correct User: {prompt}<|end_of_turn|>GPT4 Correct Assistant:"
response_text = generate_response(single_turn_prompt)
print("Response:", response_text)

## Multi-turn conversation
prompt = "Hello"
follow_up_question =  "How are you today?"
response = ""
multi_turn_prompt = f"GPT4 Correct User: {prompt}<|end_of_turn|>GPT4 Correct Assistant: {response}<|end_of_turn|>GPT4 Correct User: {follow_up_question}<|end_of_turn|>GPT4 Correct Assistant:"
response_text = generate_response(multi_turn_prompt)
print("Multi-turn conversation response:", response_text)

### Coding conversation
prompt = "Implement quicksort using C++"
coding_prompt = f"Code User: {prompt}<|end_of_turn|>Code Assistant:"
response = generate_response(coding_prompt)
print("Coding conversation response:", response)

License

The dataset, model and online demo is a research preview intended for non-commercial use only, subject to the data distillation License of LLaMA, Terms of Use of the data generated by OpenAI, and Privacy Practices of ShareGPT. Please contact us if you find any potential violation.

Acknowledgment

We would like to thank Wei-Lin Chiang from Berkeley for detailed feedback of the blog and the projects. We would like to thank the LMSYS Organization for their support of lmsys-chat-1M dataset, evaluation and online demo. We would like to thank the open source community for their efforts in providing the datasets and base models we used to develope the project, including but not limited to Anthropic, Llama, Mistral, Hugging Face H4, LMSYS, OpenChat, OpenBMB, Flan and ShareGPT.

Citation

@misc{starling2023,
    title = {Starling-7B: Improving LLM Helpfulness & Harmlessness with RLAIF},
    url = {},
    author = {Zhu, Banghua and Frick, Evan and Wu, Tianhao and Zhu, Hanlin and Jiao, Jiantao},
    month = {November},
    year = {2023}
}

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	59.92
AI2 Reasoning Challenge (25-Shot)	61.26
HellaSwag (10-Shot)	81.99
MMLU (5-Shot)	61.50
TruthfulQA (0-shot)	41.53
Winogrande (5-shot)	78.06
GSM8k (5-shot)	35.18