Update README.md

README.md

@@ -134,7 +134,7 @@ The following training hyperparameters are used in feature alignment and task sp
 
 | Data size    | Global Batch Size | Learning Rate | Epochs | Max Length | Weight Decay |
 |--------------|-------------------|---------------|--------|------------|--------------|
-| 1B        | 16               | 5e-5          | 1      | 2048       | 1e-5            |
+| 1M        | 16               | 5e-5          | 1      | 2048       | 1e-5            |
 
 We use full-parameter finetuning for the projector and apply LoRA to the language model.