[
{"prompt": "old man"}, 
{"prompt": "old male photo"}, 
{"prompt": "hairy old muscular man, realistic photography,"}
]

above are the training prompts used.

the dataset are images generated from midjourney of an ai man.

i've been using the same ai man generated from midjourney all the while so its kinda plastic looking.











{
  "engine": "kohya",
  "unetLR": 0.0005,
  "clipSkip": 1,
  "loraType": "lora",
  "keepTokens": 3,
  "networkDim": 32,
  "numRepeats": 18,
  "resolution": 1024,
  "lrScheduler": "cosine",
  "minSnrGamma": 0,
  "noiseOffset": 0.03,
  "targetSteps": 9918,
  "enableBucket": true,
  "networkAlpha": 32,
  "optimizerType": "Prodigy",
  "textEncoderLR": 0.00005,
  "maxTrainEpochs": 19,
  "shuffleCaption": true,
  "trainBatchSize": 5,
  "flipAugmentation": true,
  "lrSchedulerNumCycles": 3
}

History

07/15/2025 12:57:21 AM

Submitted

07/15/2025 12:57:21 AM

Processing

07/15/2025 10:42:30 AM

Ready

Files

145

Labels

145

Label Type

tag

Base Model

Pony