Sign In

TRAINING ERROR

83

Download

1 variant available

SafeTensor

217.87 MB

Verified:

Type
LoRA
Stats

69

83

48

Reviews
Published

Jul 19, 2024

Base Model

SDXL 1.0

Training
Epochs: 1
Usage Tips
Clip Skip: 1
Strength: 1
Hash
AutoV2
9694D0E9F7
default creator card background decoration
Followers - 538

538

Likes - 1909

1.9K

Failed Generation Contest Participant

Created on Civitai

TRAINING_ERROR_e000001_02_20240719171213.png

So what happened was I tried to train a LoRA while Civitai was having site lag issues. All the epoch sample images were multicolored noise, and the LoRAs did nothing. So I trained a LoRA using a bunch of the epoch sample images.

This is the first epoch, which showed some variety in the noise. At weights under 0.75 actual images might appear.