Commit
•
1f06411
1
Parent(s):
d9672c9
Update README.md
Browse files
README.md
CHANGED
|
@@ -55,10 +55,18 @@ pipeline_tag: text-generation
|
|
| 55 |
- The fine-tuned models demonstrate strong adaptability, excelling in human-annotated blind tests.
|
| 56 |
- The long-chat version supports extremely long texts, extending up to 200K tokens.
|
| 57 |
- The quantized versions reduce model size by 70%, improve inference speed by 30%, with performance loss less than 1%.
|
| 58 |
-
<div align="center">
|
| 59 |
-
<img src="./assets/imgs/model_cap_en.png" alt="model_cap_en" width="50%" />
|
| 60 |
-
</div>
|
| 61 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 62 |
- Orion-14B series models including:
|
| 63 |
- **Orion-14B-Base:** A multilingual large language foundational model with 14 billion parameters, pretrained on a diverse dataset of 2.5 trillion tokens.
|
| 64 |
- **Orion-14B-Chat:** A chat-model fine-tuned on a high-quality corpus aims to provide an excellence interactive experience for users in the large model community.
|
|
|
|
| 55 |
- The fine-tuned models demonstrate strong adaptability, excelling in human-annotated blind tests.
|
| 56 |
- The long-chat version supports extremely long texts, extending up to 200K tokens.
|
| 57 |
- The quantized versions reduce model size by 70%, improve inference speed by 30%, with performance loss less than 1%.
|
|
|
|
|
|
|
|
|
|
| 58 |
|
| 59 |
+
<table style="border-collapse: collapse; width: 100%;">
|
| 60 |
+
<tr>
|
| 61 |
+
<td style="border: none; padding: 10px; box-sizing: border-box;">
|
| 62 |
+
<img src="./assets/imgs/opencompass_en.png" alt="opencompass" style="width: 100%; height: auto;">
|
| 63 |
+
</td>
|
| 64 |
+
<td style="border: none; padding: 10px; box-sizing: border-box;">
|
| 65 |
+
<img src="./assets/imgs/model_cap_en.png" alt="modelcap" style="width: 100%; height: auto;">
|
| 66 |
+
</td>
|
| 67 |
+
</tr>
|
| 68 |
+
</table>
|
| 69 |
+
|
| 70 |
- Orion-14B series models including:
|
| 71 |
- **Orion-14B-Base:** A multilingual large language foundational model with 14 billion parameters, pretrained on a diverse dataset of 2.5 trillion tokens.
|
| 72 |
- **Orion-14B-Chat:** A chat-model fine-tuned on a high-quality corpus aims to provide an excellence interactive experience for users in the large model community.
|