Update README.md

README.md

@@ -102,7 +102,11 @@ In general, the way to think about these (non-orthogonal) "multiplicative-LoRAs"
 
 So instead of having just a single vector that we add (and in essence adding a bias weight to create an [affine transformation](https://en.wikipedia.org/wiki/Affine_transformation)), we now have many different control vectors that can be added (stored in `lora_B`), based on how well they match another set of "directional detection vectors" (stored in `lora_A`).
 
-**NOTE**: The [LoRA+](https://arxiv.org/abs/2402.12354) paper uses a similar way of viewing the purpose of `lora_A` and `lora_B`, but where `lora_A` looks at the ***input*** to the transformation (for "additive-LoRAs"); instead of the ***output*** of the (`down_proj`) transformation like these new (non-orthogonal) "multiplicative-LoRAs"...
+**NOTE**: The [LoRA+](https://arxiv.org/abs/2402.12354) paper uses a similar way of viewing the purpose of `lora_A` and `lora_B` (for "additive-LoRAs"):
+
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/65995c45539c808e84c38bf1/vZ2Gys3huKAWIVe0wz2-q.png)
+
+but where `lora_A` looks at the ***input*** to the transformation; instead of the ***output*** of the (`down_proj`) transformation like these new (non-orthogonal) "multiplicative-LoRAs":
 
 ---