Update README.md
Browse files
README.md
CHANGED
|
@@ -29,5 +29,6 @@ SmallThinker is designed for the following use cases:
|
|
| 29 |
Please be aware of the following limitations:
|
| 30 |
|
| 31 |
* **Language Limitation:** The model has only been trained on English-language datasets, hence its capabilities in other languages are still lacking.
|
|
|
|
| 32 |
* **Unpredictable Outputs:** The model may produce unexpected outputs due to its size and probabilistic generation paradigm. Users should exercise caution and validate the model's responses.
|
| 33 |
* **Repetition Issue:** The model tends to repeat itself when answering high-difficulty questions. Please increase the `repetition_penalty` to mitigate this issue.
|
|
|
|
| 29 |
Please be aware of the following limitations:
|
| 30 |
|
| 31 |
* **Language Limitation:** The model has only been trained on English-language datasets, hence its capabilities in other languages are still lacking.
|
| 32 |
+
* **Limited Knowledge:** Due to limited SFT data and the model's relatively small scale, its reasoning capabilities are constrained by its knowledge base.
|
| 33 |
* **Unpredictable Outputs:** The model may produce unexpected outputs due to its size and probabilistic generation paradigm. Users should exercise caution and validate the model's responses.
|
| 34 |
* **Repetition Issue:** The model tends to repeat itself when answering high-difficulty questions. Please increase the `repetition_penalty` to mitigate this issue.
|