Overbaked Lora is stored here.They are still effective at lower strength.Prompt :<pre><code>masterpiece, best quality, amazing quality, </code></pre>Optional Prompt :<pre><code>2024, 2023, 2022, </code></pre>(2024～？)Basically, the style Lora I trained contains year tags, so they are effective, but there are few year tags in some data...So the effect may not be very good. But some styles should be greatly influenced by the year tag.Negative prompt :<pre><code>url, artistname, username, text, English text, bad quality, worst quality, worst detail, sketch, censor,</code></pre>This is the recommended prompt in <a target="_blank" rel="ugc" href="https://civarchive.com/models/827184?modelVersionId=1068947">wai-NSFW-Illusory-SDXL</a>. Although my style Lora is based on <a target="_blank" rel="ugc" href="https://civarchive.com/models/795765?modelVersionId=889818">Illustrious XL</a>, I mainly use WAI model.

overbaking lora and failed work

zaflep_IL-000061.safetensors

Files with SHA256: 49d72b9b45ec69456ade7146e3378ad1436c6fcfb7170ce45eac3cffaef37118