plotting da histograms (partial snippet from Jupyter notebook)

histogram.py

import matplotlib.pyplot as plt

# …

latents: FloatTensor = self.inner_model(x, sigma, cond=cond, **kwargs)
unscaled: Tensor = latents / self.scale_factor
chs = [torch.histogram(c) for c in unscaled[0].flatten(1)]
h = torch.histogram(unscaled[0].ravel())
plt.figure(figsize=(10,2))
plt.title('Per-channel latent values after denoising sigma %.3f at CFG scale %d' % (sigma.item(), cfg_scale))
for ch, col in zip(chs, ('red','green','blue','purple',)):
	plt.hist(ch.bin_edges[:-1].cpu(), ch.bin_edges.cpu(), weights=ch.hist.cpu(), color = col, alpha = 0.4)
plt.xlabel('Latent value ÷ 0.18215')
plt.ylabel('Count')
plt.legend(['Ch0','Ch1','Ch2','Ch3'])
plt.show()

plt.figure(figsize=(10,2))
plt.title('Global latent values after denoising sigma %.3f at CFG scale %d' % (sigma.item(), cfg_scale))
plt.hist(h.bin_edges[:-1].cpu(), h.bin_edges.cpu(), weights=h.hist.cpu())
plt.xlabel('Latent value ÷ 0.18215')
plt.ylabel('Count')
plt.legend(['All'])
plt.show()

produces output such as: