can not generate normal image with pretrained model

import torch
from dalle2_pytorch import DiffusionPrior, DiffusionPriorNetwork, OpenAIClipAdapter
from dalle2_pytorch import Unet, Decoder,DALLE2

prior_network = DiffusionPriorNetwork(
    dim=768,
    depth=12,
    dim_head=64,
    heads=12,
    normformer=True,
    attn_dropout=5e-2,
    ff_dropout=5e-2,
    num_time_embeds=1,
    num_image_embeds=1,
    num_text_embeds=1,
    num_timesteps=1000,
    ff_mult=4,
    final_proj= True,
    rotary_emb= True
)

diffusion_prior = DiffusionPrior(
    net=prior_network,
    clip=OpenAIClipAdapter("ViT-L/14"),
    image_embed_dim=768,
    timesteps=1000,
    # sample_timesteps = 64,
    cond_drop_prob=0.1,
    loss_type="l2",
    condition_on_text_encodings=True,

)

diffusion_prior.load_state_dict(torch.load("prior.pth",map_location=torch.device('cpu')),strict=False)

unet = Unet(
    **{"dim": 320,
    "cond_dim": 512,
    "image_embed_dim": 768,
    "text_embed_dim": 768,
    "cond_on_text_encodings": True,
    "channels": 3,
    "dim_mults": [1, 2, 3, 4],
    "num_resnet_blocks": 4,
    "attn_heads": 8,
    "attn_dim_head": 64,
    "sparse_attn": True,
    "memory_efficient": True,
    "self_attn": [False, True, True, True]}
)

decoder = Decoder(
    unet = unet,
    clip=OpenAIClipAdapter("ViT-L/14"),
    timesteps = 1000,
    image_sizes = [64],
    image_cond_drop_prob = 0.1,
    text_cond_drop_prob = 0.5,
    learned_variance=True
)
decoder.load_state_dict(torch.load("decoder.pth",map_location=torch.device('cpu')),strict=False)

dalle2 = DALLE2(
    prior = diffusion_prior,
    decoder = decoder
)

images = dalle2(
    ['cute puppy chasing after a squirrel'],
    cond_scale = 2., # classifier free guidance strength (> 1 would strengthen the condition)
    return_pil_images=True,
)
for img in images:
    img.save("out.jpg")