Update README.md
Browse files
README.md
CHANGED
|
@@ -44,7 +44,7 @@ We provide detailed parameters and environment configurations so that you can ru
|
|
| 44 |
#### Transformers model load arguments
|
| 45 |
torch_dtype=torch.bfloat16<br>
|
| 46 |
attn_implementation='sdpa'<br>
|
| 47 |
-
**NOTE:** The
|
| 48 |
#### Instruction Adding Rules
|
| 49 |
Details can be found on our <a href="https://github.com/Kingsoft-LLM/QZhou-Embedding">GitHub</a>.
|
| 50 |
#### Evaluation code usage
|
|
|
|
| 44 |
#### Transformers model load arguments
|
| 45 |
torch_dtype=torch.bfloat16<br>
|
| 46 |
attn_implementation='sdpa'<br>
|
| 47 |
+
**NOTE:** The leaderboard evaluation results were obtained using "sdpa" mode. Other modes ('eager', 'flash_attention_2') may vary in results, but still keep the overall performance consistent.
|
| 48 |
#### Instruction Adding Rules
|
| 49 |
Details can be found on our <a href="https://github.com/Kingsoft-LLM/QZhou-Embedding">GitHub</a>.
|
| 50 |
#### Evaluation code usage
|