256x256_diffusion_uncond.pt is trained under what parameters?

[xiximelon]

[zyinghua]

I guess (under the section of classifier guidance where they use it for sampling): MODEL_FLAGS="--attention_resolutions 32,16,8 --class_cond False --diffusion_steps 1000 --image_size 256 --learn_sigma True --noise_schedule linear --num_channels 256 --num_head_channels 64 --num_res_blocks 2 --resblock_updown True --use_fp16 True --use_scale_shift_norm True"