Trigger words explaination
ataleftin: the person on the left penetrating urethra in ass to ass position
atarightin: the person on the right penetrating urethra in ass to ass position
rsbdownin: the person back to viewer with urethra being penetrated
rsfdownin: the person facing viewer with urethra being penetrated
leftin: the person on the left penetrating urethra
povin: pov penetrating urethra
rightin: the person on the right penetrating urethra
rsfupin: the person facing viewer penetrating urethra in reverse cowgirl position
upin: the person on top penetrating urethra in cowgirl position
downin: the person on back penetrating urethra in cowgirl position
tuckedleftin: the person on the left penetrating urethra on tucked penis
tuckedrightin: the person the right penetrating urethra on tucked penis
Ver Anima V1
Trained base on Anima V0.1 base with LoRA_Easy_Training_Scripts. See about this version for training details. Anima is doing awful on this concept even with this lora. So be patient. See the examples first.
Ver NOOB V-PRED V2 EQ
Trained with improved parameters based on NOOB V-PRED EQ VAE exprimental version: https://civarchive.com/models/1858821/noobai-v-pred-10-with-eq-vae
In this version, I subdivided the trigger words for more accurate generate. 12 triggerwords set. See the examples first.
Ver Noob V-PRED V1
Trained based on NOOB E-PRED V1.0 with OneTrainer. See about this version for training details. Change keywords in this version. See the examples for details.
Pros:
Better artstyle and trigger words response.
Cons:
Still a lots of trails needed
Frequent distortions of anatomy.
Ver Noob Exprimental
Trained based on NOOB E-PRED V0.75
Pros:
Better artstyle.
Cons:
Less sensitive to trigger words compared to previous versions.
Frequent distortions of anatomy.
Ver1.5 Updated
Four views set. Divided to penetrating and penetrated to prevent the twisted penis. However not sensitive to trigger words still. Also the x-ray prompt.
Used regularization techniques to reduce the influence on art style.
Some keywords may lead to overbaking, in which case decrease the trigger word weight or Lora weight.
Introduction
A Pony based LyCORIS to generate the concept Urethral intercourse. Three trigger words set. See the examples first.
Pros
Concept generation
Cons
Not responsive to trigger words, you need to upweight it or roll for it times to times
Many trials needed
Strong influence on art style
Only side view supported, back view etc will result in chaos
Training details
See about the version.
Description
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/penis_in_penis_V2/10_ataleftin"
keep_tokens = 1
name = "10"
num_repeats = 10
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/penis_in_penis_V2/10_atarightin"
keep_tokens = 1
name = "10_1790613322220391400"
num_repeats = 10
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/penis_in_penis_V2/10_rsbdownin"
keep_tokens = 1
name = "10_1790613322235903000"
num_repeats = 10
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/penis_in_penis_V2/5_leftin"
keep_tokens = 1
name = "5"
num_repeats = 5
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/penis_in_penis_V2/5_povin"
keep_tokens = 1
name = "5_1790613322266066800"
num_repeats = 5
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/penis_in_penis_V2/5_rightin"
keep_tokens = 1
name = "5_1790613322280564800"
num_repeats = 5
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/penis_in_penis_V2/5_rsfupin"
keep_tokens = 1
name = "5_1790613322295567500"
num_repeats = 5
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/penis_in_penis_V2/5_upin"
keep_tokens = 1
name = "5_1790613322310570400"
num_repeats = 5
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/penis_in_penis_V2/8_downin"
keep_tokens = 1
name = "8"
num_repeats = 8
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/penis_in_penis_V2/8_tuckedleftin"
keep_tokens = 1
name = "8_1790613322341059800"
num_repeats = 8
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/penis_in_penis_V2/8_tuckedrightin"
keep_tokens = 1
name = "8_1790613322356062700"
num_repeats = 8
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/penis_in_penis_V2/9_rsfdownin"
keep_tokens = 1
name = "9"
num_repeats = 9
random_crop_padding_percent = 0.05
shuffle_caption = true
[train_mode]
train_mode = "lora"
[general_args.args]
persistent_data_loader_workers = true
vae_batch_size = 5
pretrained_model_name_or_path = ""
mixed_precision = "bf16"
gradient_checkpointing = true
gradient_accumulation_steps = 1
seed = 42
max_data_loader_n_workers = 1
max_token_length = 225
prior_loss_weight = 1.0
sdpa = true
max_train_epochs = 20
cache_latents = true
cache_latents_to_disk = true
[general_args.dataset_args]
resolution = 1024
batch_size = 5
[network_args.args]
network_dim = 32
network_alpha = 16.0
min_timestep = 0
max_timestep = 1000
network_train_unet_only = true
[optimizer_args.args]
optimizer_type = "ProdigyPlusScheduleFree"
lr_scheduler = "constant"
loss_type = "l2"
learning_rate = 1.0
unet_lr = 1.0
max_grad_norm = 1.0
[saving_args.args]
output_dir = "G:/LoRA_Easy_Training_Scripts/output"
output_name = "penis_in_penis-anima1.0-V1-loha-dim32conv16AR0.5-NewTag"
save_precision = "bf16"
save_model_as = "safetensors"
save_every_n_epochs = 1
save_toml = true
save_toml_location = "G:/LoRA_Easy_Training_Scripts/output"
[logging_args.args]
log_prefix_mode = "disabled"
run_name_mode = "default"
[anima_args.args]
pretrained_model_name_or_path = "G:/sd-webui-forge-neo/models/Stable-diffusion/ANIMA/anima_baseV10.safetensors"
qwen3 = "G:/sd-webui-forge-neo/models/text_encoder/qwen_3_06b_base.safetensors"
vae = "G:/sd-webui-forge-neo/models/VAE/qwen_image_vae.safetensors"
qwen3_max_token_length = 512
t5_max_token_length = 512
timestep_sampling = "sigmoid"
sigmoid_scale = 1.0
[edm_loss_args.args]
edm2_loss_weighting = false
[bucket_args.dataset_args]
enable_bucket = true
min_bucket_reso = 256
max_bucket_reso = 2048
bucket_reso_steps = 64
[network_args.args.network_args]
conv_dim = 16
conv_alpha = 8.0
algo = "loha"
train_llm_adapter = "false"
[optimizer_args.args.optimizer_args]
betas = "0.95,0.99"
beta3 = "None"
weight_decay = "0"
weight_decay_by_lr = "True"
d0 = "2e-5"
d_coef = "1"
d_limiter = "True"
schedulefree_c = "0"
eps = "None"
split_groups = "True"
split_groups_mean = "False"
factored = "True"
factored_fp32 = "True"
use_bias_correction = "True"
use_stableadamw = "True"
use_schedulefree = "True"
use_speed = "False"
stochastic_rounding = "True"
fused_back_pass = "False"
use_cautious = "False"
use_grams = "False"
use_adopt = "False"
use_orthograd = "False"
use_focus = "False"