Replace links with ones from Wayback Machine (#2)
Browse files- Replace links with ones from Wayback Machine (33e01d6e96953be61b695385beaa1b28050caf31)
Co-authored-by: xzuyn <xzuyn@users.noreply.huggingface.co>
README.md
CHANGED
@@ -26,7 +26,7 @@ and consistently outperforms all the existing state-of-the-art opensource models
|
|
26 |
- WizardLM-2 70B reaches top-tier reasoning capabilities and is the first choice in the same size.
|
27 |
- WizardLM-2 7B is the fastest and achieves comparable performance with existing 10x larger opensource leading models.
|
28 |
|
29 |
-
For more details of WizardLM-2 please read our [release blog post](https://wizardlm.github.io/WizardLM2) and upcoming paper.
|
30 |
|
31 |
|
32 |
## Model Details
|
@@ -37,7 +37,7 @@ For more details of WizardLM-2 please read our [release blog post](https://wiza
|
|
37 |
* **Base model**: [mistral-community/Mixtral-8x22B-v0.1](https://huggingface.co/mistral-community/Mixtral-8x22B-v0.1)
|
38 |
* **Parameters**: 141B
|
39 |
* **Language(s)**: Multilingual
|
40 |
-
* **Blog**: [Introducing WizardLM-2](https://wizardlm.github.io/WizardLM2)
|
41 |
* **Repository**: [https://github.com/nlpxucan/WizardLM](https://github.com/nlpxucan/WizardLM)
|
42 |
* **Paper**: WizardLM-2 (Upcoming)
|
43 |
* **License**: Apache2.0
|
@@ -53,7 +53,7 @@ The WizardLM-2 8x22B even demonstrates highly competitive performance compared t
|
|
53 |
Meanwhile, WizardLM-2 7B and WizardLM-2 70B are all the top-performing models among the other leading baselines at 7B to 70B model scales.
|
54 |
|
55 |
<p align="center" width="100%">
|
56 |
-
<a ><img src="https://
|
57 |
</p>
|
58 |
|
59 |
|
@@ -67,7 +67,7 @@ We report the win:loss rate without tie:
|
|
67 |
- WizardLM-2 7B is comparable with Qwen1.5-32B-Chat, and surpasses Qwen1.5-14B-Chat and Starling-LM-7B-beta.
|
68 |
|
69 |
<p align="center" width="100%">
|
70 |
-
<a ><img src="https://
|
71 |
</p>
|
72 |
|
73 |
|
@@ -75,10 +75,10 @@ We report the win:loss rate without tie:
|
|
75 |
|
76 |
|
77 |
## Method Overview
|
78 |
-
We built a **fully AI powered synthetic training system** to train WizardLM-2 models, please refer to our [blog](https://wizardlm.github.io/WizardLM2) for more details of this system.
|
79 |
|
80 |
<p align="center" width="100%">
|
81 |
-
<a ><img src="https://
|
82 |
</p>
|
83 |
|
84 |
|
|
|
26 |
- WizardLM-2 70B reaches top-tier reasoning capabilities and is the first choice in the same size.
|
27 |
- WizardLM-2 7B is the fastest and achieves comparable performance with existing 10x larger opensource leading models.
|
28 |
|
29 |
+
For more details of WizardLM-2 please read our [release blog post](https://web.archive.org/web/20240415221214/https://wizardlm.github.io/WizardLM2/) and upcoming paper.
|
30 |
|
31 |
|
32 |
## Model Details
|
|
|
37 |
* **Base model**: [mistral-community/Mixtral-8x22B-v0.1](https://huggingface.co/mistral-community/Mixtral-8x22B-v0.1)
|
38 |
* **Parameters**: 141B
|
39 |
* **Language(s)**: Multilingual
|
40 |
+
* **Blog**: [Introducing WizardLM-2](https://web.archive.org/web/20240415221214/https://wizardlm.github.io/WizardLM2/)
|
41 |
* **Repository**: [https://github.com/nlpxucan/WizardLM](https://github.com/nlpxucan/WizardLM)
|
42 |
* **Paper**: WizardLM-2 (Upcoming)
|
43 |
* **License**: Apache2.0
|
|
|
53 |
Meanwhile, WizardLM-2 7B and WizardLM-2 70B are all the top-performing models among the other leading baselines at 7B to 70B model scales.
|
54 |
|
55 |
<p align="center" width="100%">
|
56 |
+
<a ><img src="https://web.archive.org/web/20240415175608im_/https://wizardlm.github.io/WizardLM2/static/images/mtbench.png" alt="MTBench" style="width: 96%; min-width: 300px; display: block; margin: auto;"></a>
|
57 |
</p>
|
58 |
|
59 |
|
|
|
67 |
- WizardLM-2 7B is comparable with Qwen1.5-32B-Chat, and surpasses Qwen1.5-14B-Chat and Starling-LM-7B-beta.
|
68 |
|
69 |
<p align="center" width="100%">
|
70 |
+
<a ><img src="https://web.archive.org/web/20240415163303im_/https://wizardlm.github.io/WizardLM2/static/images/winall.png" alt="Win" style="width: 96%; min-width: 300px; display: block; margin: auto;"></a>
|
71 |
</p>
|
72 |
|
73 |
|
|
|
75 |
|
76 |
|
77 |
## Method Overview
|
78 |
+
We built a **fully AI powered synthetic training system** to train WizardLM-2 models, please refer to our [blog](https://web.archive.org/web/20240415221214/https://wizardlm.github.io/WizardLM2/) for more details of this system.
|
79 |
|
80 |
<p align="center" width="100%">
|
81 |
+
<a ><img src="https://web.archive.org/web/20240415163303im_/https://wizardlm.github.io/WizardLM2/static/images/exp_1.png" alt="Method" style="width: 96%; min-width: 300px; display: block; margin: auto;"></a>
|
82 |
</p>
|
83 |
|
84 |
|