Egg Fusion - LoRa Merge

基于 512dim lora 和 chilloutmix_NiPrunedFp32Fix + deliberate_v2 的全新重制检查点。

训练数据：512 DIM LORA
/model/41893/lora-eggeaster-fusion

以下是将该 lora 应用于模型之上的效果展示：

lora 权重：TEnc Weight 0.2 UNet Weight 1

将 512dim lora 合并至 chillout 和 deliberate / --ratios 0.99 指向我的融合 lora

C 代表 CHillout / d 代表 deliberate

使用以下参数合并 C 和 D：

Base_alpha=0.53
Weight_values=0,0.157576195987654,0.28491512345679,0.384765625,0.459876543209877,0.512996720679012,0.546875,0.564260223765432,0.567901234567901,0.560546875,0.544945987654321,0.523847415123457,0.5,0.476152584876543,0.455054012345679,0.439453125,0.432098765432099,0.435739776234568,0.453125,0.487003279320987,0.540123456790124,0.615234375,0.71508487654321,0.842423804012347,1

以及
1,0.842423804012346,0.71508487654321,0.615234375,0.540123456790123,0.487003279320988,0.453125,0.435739776234568,0.432098765432099,0.439453125,0.455054012345679,0.476152584876543,0.5,0.523847415123457,0.544945987654321,0.560546875,0.567901234567901,0.564260223765432,0.546875,0.512996720679013,0.459876543209876,0.384765625,0.28491512345679,0.157576195987653,0

两者合并结果

Weight_values=
1,0.842423804012346,0.71508487654321,0.615234375,0.540123456790123,0.487003279320988,0.453125,0.435739776234568,0.432098765432099,0.439453125,0.455054012345679,0.476152584876543,0.5,0.523847415123457,0.544945987654321,0.560546875,0.567901234567901,0.564260223765432,0.546875,0.512996720679013,0.459876543209876,0.384765625,0.28491512345679,0.157576195987653,0

大功告成！
这是搭载了 512 维训练 lora 的 Egg_fusion 检查点

尽情享用你的蛋 🤟 🥃