Hugging Face's logo

Ray2333
/
GRM-Llama3.2-3B-rewardmodel-ft

Text Classification
Safetensors
llama
Model card Files Community
1
Ray2333 commited on
Commit
2780666
•
1 Parent(s): 6630d95

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +9 -8
README.md CHANGED
@@ -22,22 +22,23 @@ We evaluate GRM-Llama3.2-3B-rewardmodel-ft on the [reward model benchmark](https
22
 
23
  | Model | Average | Chat | Chat Hard | Safety | Reasoning |
24
  |:-------------------------:|:-------------:|:---------:|:---------:|:--------:|:-----------:|
25
- |[Ray2333/GRM-Llama3.2-3B-rewardmodel-ft](https://huggingface.co/Ray2333/GRM-Llama3.2-3B-rewardmodel-ft)**(ours, 3B)**|90.9|91.6|84.9|92.7|94.6|
26
- | [Ray2333/GRM-gemma2-2B-rewardmodel-ft](https://huggingface.co/Ray2333/GRM-gemma2-2B-rewardmodel-ft) **(Ours, 2B)**| 88.4 | 93.0 | 77.2 | 92.2 | 91.2 |
 
27
  | google/gemini-1.5-pro-0514 | 88.2 | 92.3 | 80.6 | 87.9 |92.0 |
28
  |RLHFlow/pair-preference-model-LLaMA3-8B |87.1 | 98.3 | 65.8|89.7|94.7|
29
- |[Ray2333/GRM-llama3-8B-sftreg](https://huggingface.co/Ray2333/GRM-llama3-8B-sftreg)**(ours, 8B)**|87.0|98.6|67.8|89.2|92.3|
30
  |google/gemini-1.5-pro-0924 | 86.8 | 94.1|77.0|85.8 |90.2|
31
  |openai/gpt-4o-2024-08-06 | 86.7 | 96.1 | 76.1 | 88.1 | 86.6|
32
- |[Ray2333/GRM-llama3.2-3B-sftreg](https://huggingface.co/Ray2333/GRM-llama3.2-3B-sftreg)**(ours, 3B)**|85.8|96.4|67.1|88.2|91.6|
33
- |[Ray2333/GRM-Gemma-2B-rewardmodel-ft](https://huggingface.co/Ray2333/GRM-Gemma-2B-rewardmodel-ft) **(Ours, 2B)**| 84.7 | 89.4 | 75.2 | 85.5 | 88.8 |
34
  | openai/gpt-4o-2024-05-13 | 84.6| 96.6 | 70.4 | 86.5 | 84.9 |
35
  | sfairXC/FsfairX-LLaMA3-RM-v0.1 (8B) | 84.4 | 99.4 | 65.1 | 86.8 | 86.4 |
36
  | Nexusflow/Starling-RM-34B | 82.6 |96.9 |57.2 |87.7 |88.5|
37
- | [Ray2333/GRM-Gemma2-2B-sftreg](https://huggingface.co/Ray2333/GRM-Gemma2-2B-sftreg)**(Ours, 2B)** | 81.0 | 97.2 | 59.6 | 86.9 | 80.3 |
38
- | [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 |
39
  | berkeley-nest/Starling-RM-7B-alpha (7B) | 74.6 | 98 | 43.4 | 88.6 | 74.6 |
40
- | [Ray2333/Gemma-2B-rewardmodel-baseline](https://huggingface.co/Ray2333/Gemma-2B-rewardmodel-baseline)**(Ours, 2B)** | 73.7 | 94.1 | 46.1 | 79.6 | 75.0 |
41
  | openbmb/UltraRM-13b (13B) | 71.3 | 96.1 | 55.3 | 45.8 | 82 |
42
 
43
 
 
22
 
23
  | Model | Average | Chat | Chat Hard | Safety | Reasoning |
24
  |:-------------------------:|:-------------:|:---------:|:---------:|:--------:|:-----------:|
25
+ |[GRM-Llama3-8B-rewardmodel-ft](https://huggingface.co/Ray2333/GRM-Llama3-8B-rewardmodel-ft)**(8B)**|91.5|95.5|86.2|90.8|93.6|
26
+ |[GRM-Llama3.2-3B-rewardmodel-ft](https://huggingface.co/Ray2333/GRM-Llama3.2-3B-rewardmodel-ft)**(ours, 3B)**|90.9|91.6|84.9|92.7|94.6|
27
+ | [GRM-gemma2-2B-rewardmodel-ft](https://huggingface.co/Ray2333/GRM-gemma2-2B-rewardmodel-ft) **(Ours, 2B)**| 88.4 | 93.0 | 77.2 | 92.2 | 91.2 |
28
  | google/gemini-1.5-pro-0514 | 88.2 | 92.3 | 80.6 | 87.9 |92.0 |
29
  |RLHFlow/pair-preference-model-LLaMA3-8B |87.1 | 98.3 | 65.8|89.7|94.7|
30
+ |[GRM-llama3-8B-sftreg](https://huggingface.co/Ray2333/GRM-llama3-8B-sftreg)**(ours, 8B)**|87.0|98.6|67.8|89.2|92.3|
31
  |google/gemini-1.5-pro-0924 | 86.8 | 94.1|77.0|85.8 |90.2|
32
  |openai/gpt-4o-2024-08-06 | 86.7 | 96.1 | 76.1 | 88.1 | 86.6|
33
+ |[GRM-llama3.2-3B-sftreg](https://huggingface.co/Ray2333/GRM-llama3.2-3B-sftreg)**(ours, 3B)**|85.8|96.4|67.1|88.2|91.6|
34
+ |[GRM-Gemma-2B-rewardmodel-ft](https://huggingface.co/Ray2333/GRM-Gemma-2B-rewardmodel-ft) **(Ours, 2B)**| 84.7 | 89.4 | 75.2 | 85.5 | 88.8 |
35
  | openai/gpt-4o-2024-05-13 | 84.6| 96.6 | 70.4 | 86.5 | 84.9 |
36
  | sfairXC/FsfairX-LLaMA3-RM-v0.1 (8B) | 84.4 | 99.4 | 65.1 | 86.8 | 86.4 |
37
  | Nexusflow/Starling-RM-34B | 82.6 |96.9 |57.2 |87.7 |88.5|
38
+ | [GRM-Gemma2-2B-sftreg](https://huggingface.co/Ray2333/GRM-Gemma2-2B-sftreg)**(Ours, 2B)** | 81.0 | 97.2 | 59.6 | 86.9 | 80.3 |
39
+ | [GRM-Gemma-2B-sftreg](https://huggingface.co/Ray2333/GRM-Gemma-2B-sftreg)**(Ours, 2B)** | 75.3 | 95.5 | 48.7 | 80.0 | 76.8 |
40
  | berkeley-nest/Starling-RM-7B-alpha (7B) | 74.6 | 98 | 43.4 | 88.6 | 74.6 |
41
+ | [Gemma-2B-rewardmodel-baseline](https://huggingface.co/Ray2333/Gemma-2B-rewardmodel-baseline)**(Ours, 2B)** | 73.7 | 94.1 | 46.1 | 79.6 | 75.0 |
42
  | openbmb/UltraRM-13b (13B) | 71.3 | 96.1 | 55.3 | 45.8 | 82 |
43
 
44