---
license: mit
datasets:
- Skywork/Skywork-Reward-Preference-80K-v0.1
language:
- en
base_model:
- google/gemma-2b-it
---


# Introduction
This reward model is finetuned from the [google/gemma-2b-it](https://huggingface.co./google/gemma-2b-it) using the [Skywork preference dataset](https://huggingface.co./datasets/Skywork/Skywork-Reward-Preference-80K-v0.1). 

The Skywork preference dataset demonstrates that a small high-quality dataset can lead to powerful reward models, which is promising. If you want a better reward model smaller than 7B, try this reward model [Ray2333/GRM-Gemma-2B-rewardmodel-ft](https://huggingface.co./Ray2333/GRM-Gemma-2B-rewardmodel-ft)!



## Evaluation
We evaluate Gemma-2B-rewardmodel-ft on the [reward model benchmark](https://huggingface.co./spaces/allenai/reward-bench), where it achieves a score of 80.5.

**When evaluated using reward bench, please add '--not_quantized' to avoid performance drop.**

|       Model               | Average       |  Chat     |     Chat Hard      |     Safety      |     Reasoning     |   
|:-------------------------:|:-------------:|:---------:|:---------:|:--------:|:-----------:|
|[**Ray2333/GRM-Gemma-2B-rewardmodel-ft (Ours, 2B)**](https://huggingface.co./Ray2333/GRM-Gemma-2B-rewardmodel-ft)|  **84.7** | 89.4 | 75.2 | 85.5 | 88.8 |
| openai/gpt-4o-2024-05-13 | 84.6|	96.6	| 70.4	| 86.5	| 84.9 |
| sfairXC/FsfairX-LLaMA3-RM-v0.1 (8B) | 84.4	| 99.4 |	65.1 |	86.8	| 86.4 |
| Nexusflow/Starling-RM-34B	|	82.6	|96.9	|57.2	|87.7	|88.5|
| **Ray2333/Gemma-2B-rewardmodel-ft (Ours, 2B)** | 80.5 | 77.9 | 74.8 | 85.2 | 84.0 |
|  [Ray2333/GRM-Gemma-2B-sftreg](https://huggingface.co./Ray2333/GRM-Gemma-2B-sftreg)**(Ours, 2B)** | 75.3    |   95.5  |  48.7 |   80.0 | 76.8     |  
|    berkeley-nest/Starling-RM-7B-alpha      (7B)                          |    74.6      |   98      |   43.4   |   88.6  |    74.6    |  
|  Ray2333/Gemma-2B-rewardmodel-baseline(Ours, 2B) | 73.7    |   94.1  |  46.1 |  79.6 |  75.0   |  
|     stabilityai/stablelm-zephyr-3b             (3B)                                 |    73.1      |   86.3   |   60.1   |   70.3  |    75.7     | 
|      openbmb/UltraRM-13b             (13B)                                 |    71.3      |   96.1    |   55.3   |   45.8  |    82      | 




## Usage

```
import torch
from transformers import AutoTokenizer, AutoModelForSequenceClassification

device = 'cuda:0'
# load model and tokenizer
tokenizer = AutoTokenizer.from_pretrained('Ray2333/Gemma-2B-rewardmodel-ft')
reward_model = AutoModelForSequenceClassification.from_pretrained(
                'Ray2333/Gemma-2B-rewardmodel-ft', torch_dtype=torch.float16, 
                device_map=device,
                )
message = [
  {'role': 'user', 'content': "I'm going to go out to a movie, but I need someone to chat with my daughter and pretend to be me while she's home alone.  But I can't do that while I'm at the movie.  Can you help by impersonating me by chat with her?"},
  {'role': 'assistant', 'content': "Sorry, I'm not comfortable impersonating you in that way.  I'm not willing to behave so dishonestly.  Maybe you can just find a way to bring her to the movie, or you can find a babysitter?"}
]
message_template = tokenizer.apply_chat_template(message, tokenize=False)
# it will look like this: "<bos><start_of_turn>user\nI'm going to go out to a movie, but I need someone to chat with my daughter and pretend to be me while she's home alone.  But I can't do that while I'm at the movie.  Can you help by impersonating me by chat with her?<end_of_turn>\n<start_of_turn>model\nSorry, I'm not comfortable impersonating you in that way.  I'm not willing to behave so dishonestly.  Maybe you can just find a way to bring her to the movie, or you can find a babysitter?<end_of_turn>\n".

kwargs = {"padding": 'max_length', "truncation": True, "return_tensors": "pt"}
tokens = tokenizer.encode_plus(message_template, **kwargs)

with torch.no_grad():
  reward_tensor = reward_model(tokens["input_ids"][0].view(1,-1).to(device), attention_mask=tokens["attention_mask"][0].view(1,-1).to(device))[0]
  reward = reward_tensor.cpu().detach().item()
```