Update Model Card
Browse files
README.md
CHANGED
|
@@ -71,6 +71,11 @@ We introduce **YuLan-Mini-Instruct**, a compact yet powerful model with 2.4 bill
|
|
| 71 |
YuLan-Mini-Instruct is pre-trained on 1.08 trillion tokens and further enhanced through a post-training pipeline that incorporates Supervised Fine-Tuning, Preference Finetuning, and Reinforcement Learning to maximize its capabilities.
|
| 72 |
Extensive evaluations demonstrate that YuLan-Mini-Instruct achieves state-of-the-art performance, demonstrating competitive capabilities with leading industrial counterparts across multiple domains including instruction-following, math, code, and reasoning tasks. The model is available at https://huggingface.co/yulan-team/YuLan-Mini-Instruct.
|
| 73 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 74 |
## 2. Model Overview
|
| 75 |
|
| 76 |
**YuLan-Mini-Instruct** exhibits competitive performance compared to similarly sized models, particularly in reasoning tasks involving math and code.
|
|
|
|
| 71 |
YuLan-Mini-Instruct is pre-trained on 1.08 trillion tokens and further enhanced through a post-training pipeline that incorporates Supervised Fine-Tuning, Preference Finetuning, and Reinforcement Learning to maximize its capabilities.
|
| 72 |
Extensive evaluations demonstrate that YuLan-Mini-Instruct achieves state-of-the-art performance, demonstrating competitive capabilities with leading industrial counterparts across multiple domains including instruction-following, math, code, and reasoning tasks. The model is available at https://huggingface.co/yulan-team/YuLan-Mini-Instruct.
|
| 73 |
|
| 74 |
+
| Model | Context Length | SFT | DPO | 🤗 Hugging Face | ModelScope | Wise Model |
|
| 75 |
+
|---------|----------------|-----|-----|-----------------|------------|------------|
|
| 76 |
+
| YuLan-Mini | 28K | ❎ | ❎ | [`Base`](https://huggingface.co/yulan-team/YuLan-Mini) | [`Base`](https://modelscope.cn/models/yulan-team/YuLan-Mini) | [`Base`](https://wisemodel.cn/models/yulan-team/YuLan-Mini) |
|
| 77 |
+
| YuLan-Mini-Instruct | 28K | ✅ | ✅ | [`Instruct`](https://huggingface.co/yulan-team/YuLan-Mini-Instruct) | | |
|
| 78 |
+
|
| 79 |
## 2. Model Overview
|
| 80 |
|
| 81 |
**YuLan-Mini-Instruct** exhibits competitive performance compared to similarly sized models, particularly in reasoning tasks involving math and code.
|