Update README.md

README.md

@@ -51,10 +51,9 @@ Training was first performed using nf4 quantization for 32 epochs (8 epochs coun
 Training then continued at int8 quantization for another 16 epochs (4 epochs counting the original + augmentations as a single epoch):
 - `qwen-360-diffusion-int8-bf16-v1.safetensors` was trained for a total of 48 epochs (2,304,000 steps)
 
-Training continued at int8 quantization for another x epochs on the [Qwen/Qwen-Image-2512](https://huggingface.co/Qwen/Qwen-Image-2512) base model:
+Training then continued on [Qwen/Qwen-Image-2512](https://huggingface.co/Qwen/Qwen-Image-2512) at int8 quantization for another 14 epochs on (3.5 epochs counting the original + augmentations as a single epoch):
 - The total number of equirectangular images increased to 35k, and regularization images were removed from the dataset.
-- The loss was adjusted to be based on the information density of equirectangular images.
-- `qwen-360-diffusion-2512-int8-bf16-v2.safetensors` was trained for x epochs
+- `qwen-360-diffusion-2512-int8-bf16-v2.safetensors` was trained for 62 epochs (2,794,000 steps)
 
 ---