update README for epoch 21

README.md

@@ -9,7 +9,7 @@ base_model:
 A latent diffusion model (LDM) geared toward illustration, style composability, and sample variety. Addresses a few deficiencies with the SDXL base model; feels more like an SD 1.x with better resolution and much better prompt adherence.
 
 * Architecture: SD XL (base model is v1.0)
-* Training procedure: U-Net fully unfrozen, all-parameter continued pretraining at LR between 3e-8 and 3e-7 for 20,110,000 steps (at epoch 19, batch size 4). See below for more details.
+* Training procedure: U-Net fully unfrozen, all-parameter continued pretraining at LR between 3e-8 and 3e-7 for 22,830,000 steps (at epoch 21, batch size 4). See below for more details.
 
 Trained on the Puzzle Box dataset, a large collection of permissively licensed images from the public Internet (or generated by previous Puzzle Box models). Each image 
 has from 3 to 24 different captions which are used interchangably during training. There are approximately 12 million images and 78 million captions in the dataset.
@@ -22,8 +22,10 @@ booru-style, don't use underscores in your tags, replace those with spaces. Tags
 
 Vitamin phrases: *top quartile*, *top decile* (there are also anti-vitamins, *bottom quartile* and *bottom decile*). These are the primary aesthetic labels (see below.)
 
-Prompt adherence is unusually good; aesthetics are improved by human evaluation for generations between 1/4 and 1/2 megapixel in size for epochs 12-14, 1/4 to 2 
-megapixels for epoch 15+. CFG scales between 2 and 7 can work well with Puzzle Box; experimenting with resolution or scale for your prompts is encouraged.
+Prompt adherence is unusually good for this architecture; aesthetics are improved by human evaluation for generations between 1/4 and 1/2 megapixel in size for epochs 12-14, 1/4 to 2 
+megapixels for epoch 15+.
+
+Inference suggestions: Euler ancestral sampling, CFG between 2-5 (lower for more recent checkpoints) at 20-50 steps. Experimenting with resolution or scale for your prompts is encouraged.
 
 **Captioning:** About 1.4 million of the captions in the dataset are human-written. The remainder come from a variety of ML models, either vision transformers or 
 classifers. Models used in captioning the Puzzle Box dataset include: Qwen 2 VL 72b, BLIP 2 OPT-6.5B COCO, Llava 1.5, MiniCPM 2.6, bakllava, Moondream, DeepSeek Janus 7b,
@@ -43,7 +45,7 @@ In later epochs, a form of curriculum training is used: a complexity proxy is ca
 
 Epoch length was determined by the original size of the training set, and the best checkpoint that emerges after model soup experimentation is released.
 
-**Safety notes:** Similar to the base SD XL model, deployment in a production environment may require filters for undesired/inappropriate content. Classifiers on both input prompt and output image suggested.
+**Safety notes:** Similar to the base SD XL model, deployment in a production environment may require filters for undesired/inappropriate content. Classifiers on both input prompt and output image suggested. The model is generally heavily biased towards aesthetics and diversity of output; it may not reflect historical or population norms and its world knowledge is incomplete.
 
 **Other nifty tricks used:** Some less common techniques used in training Puzzle Box XL include:
 
@@ -56,6 +58,8 @@ Epoch length was determined by the original size of the training set, and the be
 
 Model checkpoints currently available:
 
+- from epoch 21, **22830k** training steps, 19 June 2026
+- from epoch 20, **21410k** training steps, 21 January 2026
 - from epoch 19, **20110k** training steps, 28 November 2025
 - from epoch 18, **19300k** training steps, 03 October 2025
 - from epoch 17, **18000k** training steps, 06 July 2025
@@ -74,6 +78,8 @@ interpolation is best.)
 The U-Net attention layers are the layers most modified by the continued pretrain; comparing those layers to SD XL 1.0, the correlation is:
 | Epoch |    Date    | R-squared |
 | ----- | ---------- | --------- |
+|  21   | 2026-06-19 |  96.914%  |
+|  20   | 2026-01-21 |  97.082%  |
 |  19   | 2025-11-28 |  97.257%  |
 |  18   | 2025-10-03 |  97.426%  |
 |  17   | 2025-07-06 |  97.705%  |