Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -7,10 +7,8 @@ widget:
|
|
7 |
text: hi
|
8 |
output:
|
9 |
text: ' Hello! How can I assist you today?'
|
10 |
-
|
11 |
pipeline_tag: text-generation
|
12 |
---
|
13 |
-
|
14 |
# 🐷SUS-Chat: Instruction tuning done right
|
15 |
|
16 |
<p align="left">
|
@@ -187,8 +185,7 @@ data-layout-align="center">
|
|
187 |
</tr>
|
188 |
<tr class="even">
|
189 |
<td style="text-align: right;">SUS-Chat-34B</td>
|
190 |
-
<td style="text-align: center;"><
|
191 |
-
class="math inline">$\underline{74.35}$</span></td>
|
192 |
</tr>
|
193 |
<tr class="odd">
|
194 |
<td style="text-align: right;">Qwen-72b-Chat</td>
|
@@ -240,10 +237,8 @@ role="doc-noteref"><sup>1</sup></a></th>
|
|
240 |
</tr>
|
241 |
<tr class="odd">
|
242 |
<td style="text-align: right;">Qwen-72b-Chat</td>
|
243 |
-
<td style="text-align: center;"><
|
244 |
-
|
245 |
-
<td style="text-align: center;"><span
|
246 |
-
class="math inline">$\underline{77.22}$</span></td>
|
247 |
</tr>
|
248 |
<tr class="even">
|
249 |
<td style="text-align: right;">Deepseek-68b-Chat</td>
|
@@ -280,25 +275,25 @@ role="doc-backlink">↩︎</a></p></li>
|
|
280 |
|
281 |
## Math & Reasoning
|
282 |
|
283 |
-
| Model |
|
284 |
-
|
285 |
-
| GPT-4 |
|
286 |
-
| SUS-Chat-34B |
|
287 |
-
| Qwen-72b-Chat |
|
288 |
-
| Deepseek-68b-Chat |
|
289 |
-
| OrionStar-Yi-34B-Chat |
|
290 |
-
| Yi-34B-Chat |
|
291 |
|
292 |
## More Tasks
|
293 |
|
294 |
-
| Model | winogrande (5-shot) |
|
295 |
-
|
296 |
-
| GPT-4 | — |
|
297 |
-
| SUS-Chat-34B | **81.22** |
|
298 |
-
| Qwen-72b-Chat | 76.09 |
|
299 |
-
| Deepseek-68b-Chat |
|
300 |
-
| OrionStar-Yi-34B-Chat | 77.27 |
|
301 |
-
| Yi-34B-Chat | 76.64 |
|
302 |
|
303 |
## Overall
|
304 |
|
@@ -400,4 +395,4 @@ model.
|
|
400 |
This model is developed entirely for academic research and free
|
401 |
commercial use, but it must adhere to the
|
402 |
[license](https://github.com/01-ai/Yi/blob/main/MODEL_LICENSE_AGREEMENT.txt)
|
403 |
-
from [01-ai](https://huggingface.co/01-ai).
|
|
|
7 |
text: hi
|
8 |
output:
|
9 |
text: ' Hello! How can I assist you today?'
|
|
|
10 |
pipeline_tag: text-generation
|
11 |
---
|
|
|
12 |
# 🐷SUS-Chat: Instruction tuning done right
|
13 |
|
14 |
<p align="left">
|
|
|
185 |
</tr>
|
186 |
<tr class="even">
|
187 |
<td style="text-align: right;">SUS-Chat-34B</td>
|
188 |
+
<td style="text-align: center;"><u>74.35</u></td>
|
|
|
189 |
</tr>
|
190 |
<tr class="odd">
|
191 |
<td style="text-align: right;">Qwen-72b-Chat</td>
|
|
|
237 |
</tr>
|
238 |
<tr class="odd">
|
239 |
<td style="text-align: right;">Qwen-72b-Chat</td>
|
240 |
+
<td style="text-align: center;"><u>77.02</u></td>
|
241 |
+
<td style="text-align: center;"><u>77.22</u></td>
|
|
|
|
|
242 |
</tr>
|
243 |
<tr class="even">
|
244 |
<td style="text-align: right;">Deepseek-68b-Chat</td>
|
|
|
275 |
|
276 |
## Math & Reasoning
|
277 |
|
278 |
+
| Model | gsm8k (0-shot) | MATH (0-shot) | BBH (0-shot) |
|
279 |
+
|----------------------:|:--------------:|:-------------:|:------------:|
|
280 |
+
| GPT-4 | 91.4 | 45.8 | 86.7 |
|
281 |
+
| SUS-Chat-34B | **80.06** | 28.7 | 67.62 |
|
282 |
+
| Qwen-72b-Chat | <u>76.57</u> | **35.9** | **72.63** |
|
283 |
+
| Deepseek-68b-Chat | 74.45 | <u>29.56</u> | <u>69.73</u> |
|
284 |
+
| OrionStar-Yi-34B-Chat | 54.36 | 12.8 | 62.88 |
|
285 |
+
| Yi-34B-Chat | 63.76 | 10.02 | 61.54 |
|
286 |
|
287 |
## More Tasks
|
288 |
|
289 |
+
| Model | winogrande (5-shot) | arc (25-shot) | hellaswag (10-shot) | TruthfulQA mc1 (0-shot) | TruthfulQA mc2 (0-shot) |
|
290 |
+
|----------------------:|:-------------------:|:-------------:|:-------------------:|:-----------------------:|:-----------------------:|
|
291 |
+
| GPT-4 | — | 94.5 | 91.4 | 59.00 | — |
|
292 |
+
| SUS-Chat-34B | **81.22** | <u>81.54</u> | 83.79 | **40.64** | **57.47** |
|
293 |
+
| Qwen-72b-Chat | 76.09 | **82.10** | <u>86.06</u> | 39.17 | <u>56.37</u> |
|
294 |
+
| Deepseek-68b-Chat | <u>80.58</u> | 81.29 | **87.02** | <u>40.02</u> | 50.64 |
|
295 |
+
| OrionStar-Yi-34B-Chat | 77.27 | 80.19 | 84.54 | 36.47 | 53.24 |
|
296 |
+
| Yi-34B-Chat | 76.64 | 70.66 | 82.29 | 38.19 | 54.57 |
|
297 |
|
298 |
## Overall
|
299 |
|
|
|
395 |
This model is developed entirely for academic research and free
|
396 |
commercial use, but it must adhere to the
|
397 |
[license](https://github.com/01-ai/Yi/blob/main/MODEL_LICENSE_AGREEMENT.txt)
|
398 |
+
from [01-ai](https://huggingface.co/01-ai).
|