Answer time_embedder.linear_1.weight.dtype for diffusers before #14398

Answer time_embedder.linear_1.weight.dtype, which diffusers before #14398 reads

Add truncated text encoder (decoder layers 0-50) and point the index at it

Project released-checkpoint AdaLN LoRAs onto the pruned timestep coordinates

Add opt-in ConvRot 8-bit compute (enable_convrot / quantize_8bit) + measured ladder

Model card

transformer_ref: weight index

transformer_ref: diffusion_pytorch_model-00014-of-00014.safetensors

transformer_ref: diffusion_pytorch_model-00013-of-00014.safetensors

transformer_ref: diffusion_pytorch_model-00012-of-00014.safetensors

transformer_ref: diffusion_pytorch_model-00011-of-00014.safetensors

transformer_ref: diffusion_pytorch_model-00010-of-00014.safetensors

transformer_ref: diffusion_pytorch_model-00009-of-00014.safetensors

transformer_ref: diffusion_pytorch_model-00008-of-00014.safetensors

transformer_ref: diffusion_pytorch_model-00007-of-00014.safetensors

transformer_ref: diffusion_pytorch_model-00006-of-00014.safetensors

transformer_ref: diffusion_pytorch_model-00005-of-00014.safetensors

transformer_ref: diffusion_pytorch_model-00004-of-00014.safetensors

transformer_ref: diffusion_pytorch_model-00003-of-00014.safetensors

Config, modular index and the pruned transformer class

transformer_ref: diffusion_pytorch_model-00002-of-00014.safetensors

transformer_ref: diffusion_pytorch_model-00001-of-00014.safetensors

transformer: weight index

transformer: diffusion_pytorch_model-00014-of-00014.safetensors

transformer: diffusion_pytorch_model-00013-of-00014.safetensors

transformer: diffusion_pytorch_model-00012-of-00014.safetensors

transformer: diffusion_pytorch_model-00011-of-00014.safetensors

transformer: diffusion_pytorch_model-00010-of-00014.safetensors

transformer: diffusion_pytorch_model-00009-of-00014.safetensors

transformer: diffusion_pytorch_model-00008-of-00014.safetensors

transformer: diffusion_pytorch_model-00007-of-00014.safetensors

transformer: diffusion_pytorch_model-00006-of-00014.safetensors

transformer: diffusion_pytorch_model-00005-of-00014.safetensors

transformer: diffusion_pytorch_model-00004-of-00014.safetensors

transformer: diffusion_pytorch_model-00003-of-00014.safetensors

transformer: diffusion_pytorch_model-00002-of-00014.safetensors

transformer: diffusion_pytorch_model-00001-of-00014.safetensors

initial commit