Erorr trying to use laion7 plus or sac8 plus

[shadinx2]

Traceback (most recent call last): File "C:\Users\Dimitrie\stable-diffusion-webui\modules\ui.py", line 223, in f res = list(func(*args, **kwargs)) File "C:\Users\Dimitrie\stable-diffusion-webui\webui.py", line 63, in f res = func(*args, **kwargs) File "C:\Users\Dimitrie\stable-diffusion-webui\modules\txt2img.py", line 48, in txt2img processed = process_images(p) File "C:\Users\Dimitrie\stable-diffusion-webui\modules\processing.py", line 407, in process_images uc = prompt_parser.get_learned_conditioning(shared.sd_model, len(prompts) * [p.negative_prompt], p.steps) File "C:\Users\Dimitrie\stable-diffusion-webui\modules\prompt_parser.py", line 138, in get_learned_conditioning conds = model.get_learned_conditioning(texts) File "C:\Users\Dimitrie\stable-diffusion-webui\repositories\stable-diffusion\ldm\models\diffusion\ddpm.py", line 558, in get_learned_conditioning c = self.cond_stage_model(c) File "C:\Users\Dimitrie\stable-diffusion-webui\venv\lib\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl return forward_call(*input, **kwargs) File "C:\Users\Dimitrie\stable-diffusion-webui\modules\sd_hijack.py", line 334, in forward z1 = self.process_tokens(tokens, multipliers) File "C:\Users\Dimitrie\stable-diffusion-webui\extensions\aesthetic-gradients\aesthetic_clip.py", line 219, in call img_embs = slerp(self.image_embs, text_embs_2, self.aesthetic_slerp_angle) File "C:\Users\Dimitrie\stable-diffusion-webui\extensions\aesthetic-gradients\aesthetic_clip.py", line 139, in slerp low_norm = low / torch.norm(low, dim=1, keepdim=True) File "C:\Users\Dimitrie\stable-diffusion-webui\venv\lib\site-packages\torch\functional.py", line 1472, in norm return _VF.frobenius_norm(input, _dim, keepdim=keepdim) IndexError: Dimension out of range (expected to be in range of [-1, 0], but got 1)

My own aesthetic embeddings and fantasy(from the original repo) work

[uwusensei]

this issue occurs for me too, and only on laion7 and sac8 embeddings

It goes away if you don't use the 'Aesthetic text for imgs' feature, as hinted at by the stack trace above struggling with the conditionals