This version works with SD 2.1-768

You need to specify <cartoonish_doll> in the prompt for the style to work.

Some examples of how the style is applied to an image you can see here https://www.reddit.com/r/StableDiffusion/comments/zqnw8o/my_text_inversion_model_for_sd_21768/