[cite_start]There are three different training processes, each with its own job ID, start time, and base model[cite: 1, 2, 3]. [cite_start]All three training jobs have the same number of files and labels (75 each) and use "tag" as the label type[cite: 1, 2, 3]. [cite_start]They all use the Kohya engine, a resolution of 1024, and have a <code>clipSkip</code> value of 1[cite: 1, 2, 3]. [cite_start]The <code>loraType</code> is "lora" for all of them, and <code>maxTrainEpochs</code> is set to 19[cite: 1, 2, 3]. [cite_start]All three jobs also have a <code>targetSteps</code> of 9975 and have <code>enableBucket</code>, <code>shuffleCaption</code>, and <code>flipAugmentation</code> enabled[cite: 1, 2, 3].***### Job 563759-1[cite_start]This training job started on September 8, 2025, at 6:55:14 AM and used the Pony base model[cite: 1]. [cite_start]The <code>optimizerType</code> is Prodigy, and the <code>lrScheduler</code> is cosine[cite: 1]. [cite_start]The <code>networkDim</code> is 64, and the <code>networkAlpha</code> is 32[cite: 1]. [cite_start]The <code>trainBatchSize</code> is 5, and <code>numRepeats</code> is 35[cite: 1]. [cite_start]It also has a <code>noiseOffset</code> of 0.03 and a <code>minSnrGamma</code> of 0[cite: 1].***### Job 563759-2[cite_start]This job began on September 8, 2025, at 6:56:09 AM, with SDXL as the base model[cite: 2]. [cite_start]It uses Adafactor as the <code>optimizerType</code> and cosine_with_restarts for the <code>lrScheduler</code>[cite: 2]. [cite_start]The <code>networkDim</code> is 32, and the <code>networkAlpha</code> is 16[cite: 2]. [cite_start]The <code>trainBatchSize</code> is 4, and <code>numRepeats</code> is 28[cite: 2]. [cite_start]This job has a <code>noiseOffset</code> of 0.1 and a <code>minSnrGamma</code> of 5[cite: 2].***### Job 563759-3[cite_start]Starting on September 8, 2025, at 6:54:11 AM, this training used the Illustrious base model[cite: 3]. [cite_start]Similar to the SDXL job, it uses Adafactor as the <code>optimizerType</code> and cosine_with_restarts as the <code>lrScheduler</code>[cite: 3]. [cite_start]The <code>networkDim</code> is 32, and the <code>networkAlpha</code> is 16[cite: 3]. [cite_start]The <code>trainBatchSize</code> is 4, and <code>numRepeats</code> is 28[cite: 3]. [cite_start]It also has a <code>noiseOffset</code> of 0.1 and a <code>minSnrGamma</code> of 5[cite: 3].

20250908-0647-another-midjourney-old-man

Based on the phrases used in the training files, here is a grouping of the descriptions.General Descriptors of "just1n"This group includes phrases that describe the subject's general appearance across multiple files.Physical traits: "1boy" , "male focus" , "solo" , "glasses" , "realistic" , and "looking at viewer".Facial expressions: "smile" , "teeth" , and "facial hair".Formal AttireThese phrases describe a formal event with the subject wearing a suit or tuxedo.Attire: "bowtie" , "black bowtie" , "suit" , "tuxedo" , "jacket" , "black jacket" , "formal" , and "open jacket".Setting: "indoors".Casual Attire and SettingsThis group is for phrases that describe everyday clothes and casual environments.Clothing: "shirt" , "tank top" , "sleeveless" , "shorts" , "collared shirt" , and "backpack".Context: "outdoors" , "indoors" , and "shop".Shirtless and "Bara" Body TypeThese phrases are used to describe the subject's physique and appearance when shirtless.Body type: "muscular male" , "muscular" , "bara" , "fat" , "plump" , "fat man" , "belly" , "abs" , "pectorals" , "large pectorals" , and "chest hair".State: "topless male" , "shirtless" , "bare pectorals" , and "nude".

Files with SHA256: 3bc9ffd3da041b1e5b74c861088ac24b04842b1f64f2ea769a6117cb94929a14