leran1995 commited on
Commit
bfd16d8
·
verified ·
1 Parent(s): cec014e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -4
README.md CHANGED
@@ -59,23 +59,23 @@ To ensure a fair comparison, we conducted experiments with three distinct datase
59
  |--------------------|----------|----------|----------|------|
60
  | Qwen3-4B-Base | 24.6 | 25.0 | 90.4 | 44.6 |
61
  | Qwen3-8B-Base | <u>37.9<u> | <u>29.6<u> | <u>91.1<u> | <u>48.9<u> |
62
- | **Nezhaige4-3B-Base** | **52.9** | **40.8** | **93.4** | **53.4** |
63
 
64
  * Finetuned with Ring-lite-sft-data
65
  | Model | AIME2024 | AIME2025 | Math-500 | GPQA |
66
  |--------------------|----------|----------|----------|------|
67
  | Qwen3-4B-Base | 40.4 | 31.3 | 93.6 | 51.4 |
68
  | Qwen3-8B-Base | <u>50.0<u> | <u>35.8<u> | <u>94.4<u> | <u>55.1<u> |
69
- | **Nezhaige4-3B-Base** | **56.8** | **45.3** | **95.5** | **57.7** |
70
 
71
  * Finetuned with OpenThoughts3
72
  | Model | AIME2024 | AIME2025 | Math-500 | GPQA |
73
  |--------------------|----------|----------|----------|------|
74
  | Qwen3-4B-Base | 52.9 | 42.1 | 93.2 | 49.6 |
75
  | Qwen3-8B-Base | <u>60.4<u> | <u>47.1<u> | **95.0** | <u>55.3<u> |
76
- | **Nezhaige4-3B-Base** | **62.4** | **49.2** | <u>94.6<u> | **56.9** |
77
 
78
- The results demonstrate that **Nezhaige4-3B-Base** significantly outperforms Qwen3-4B-Base, and even surpasses the larger Qwen3-8B-Base, highlighting the greater potential of our base model after fine-tuning. This advantage stems from the optimized training recipe during our Stable stage and the extensive high-quality synthetic data incorporated during the Decay stage.
79
 
80
 
81
  ## <span id="Inference">4. Quickstart</span>
 
59
  |--------------------|----------|----------|----------|------|
60
  | Qwen3-4B-Base | 24.6 | 25.0 | 90.4 | 44.6 |
61
  | Qwen3-8B-Base | <u>37.9<u> | <u>29.6<u> | <u>91.1<u> | <u>48.9<u> |
62
+ | **Nanbeige4-3B-Base** | **52.9** | **40.8** | **93.4** | **53.4** |
63
 
64
  * Finetuned with Ring-lite-sft-data
65
  | Model | AIME2024 | AIME2025 | Math-500 | GPQA |
66
  |--------------------|----------|----------|----------|------|
67
  | Qwen3-4B-Base | 40.4 | 31.3 | 93.6 | 51.4 |
68
  | Qwen3-8B-Base | <u>50.0<u> | <u>35.8<u> | <u>94.4<u> | <u>55.1<u> |
69
+ | **Nanbeige4-3B-Base** | **56.8** | **45.3** | **95.5** | **57.7** |
70
 
71
  * Finetuned with OpenThoughts3
72
  | Model | AIME2024 | AIME2025 | Math-500 | GPQA |
73
  |--------------------|----------|----------|----------|------|
74
  | Qwen3-4B-Base | 52.9 | 42.1 | 93.2 | 49.6 |
75
  | Qwen3-8B-Base | <u>60.4<u> | <u>47.1<u> | **95.0** | <u>55.3<u> |
76
+ | **Nanbeige4-3B-Base** | **62.4** | **49.2** | <u>94.6<u> | **56.9** |
77
 
78
+ The results demonstrate that **Nanbeige4-3B-Base** significantly outperforms Qwen3-4B-Base, and even surpasses the larger Qwen3-8B-Base, highlighting the greater potential of our base model after fine-tuning. This advantage stems from the optimized training recipe during our Stable stage and the extensive high-quality synthetic data incorporated during the Decay stage.
79
 
80
 
81
  ## <span id="Inference">4. Quickstart</span>