error when load checkpoints
when i load git-large, some error come up.
loading code:
git_model='/HOME/GIT/weights/git-large'
self.git_transform = AutoProcessor.from_pretrained(git_model)
self.git = GitVisionModel.from_pretrained(git_model)
error is：
Some weights of GitVisionModel were not initialized from the model checkpoint at /HOME/GIT/weights/git-large and are newly initialized: ['git.vision_model.encoder.layers.23.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.2.mlp.fc1.weight', 'git.vision_model.encoder.layers.17.layer_norm2.weight', 'git.vision_model.encoder.layers.12.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.15.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.15.layer_norm1.bias', 'git.vision_model.encoder.layers.2.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.10.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.17.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.10.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.1.layer_norm2.bias', 'git.vision_model.encoder.layers.4.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.0.layer_norm1.bias', 'git.vision_model.encoder.layers.13.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.12.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.9.mlp.fc2.bias', 'git.vision_model.encoder.layers.16.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.20.layer_norm1.bias', 'git.vision_model.encoder.layers.7.mlp.fc2.bias', 'git.vision_model.encoder.layers.12.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.18.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.20.layer_norm2.bias', 'git.vision_model.encoder.layers.1.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.22.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.19.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.12.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.10.layer_norm1.bias', 'git.vision_model.encoder.layers.19.layer_norm2.weight', 'git.vision_model.encoder.layers.11.layer_norm2.weight', 'git.vision_model.encoder.layers.9.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.22.layer_norm1.bias', 'git.vision_model.encoder.layers.17.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.18.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.23.mlp.fc2.bias', 'git.vision_model.encoder.layers.19.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.8.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.21.mlp.fc1.weight', 'git.vision_model.encoder.layers.3.layer_norm1.bias', 'git.vision_model.encoder.layers.11.mlp.fc2.weight', 'git.vision_model.encoder.layers.22.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.23.layer_norm2.weight', 'git.vision_model.encoder.layers.5.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.6.layer_norm2.weight', 'git.vision_model.pre_layrnorm.weight', 'git.vision_model.encoder.layers.19.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.13.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.10.layer_norm2.weight', 'git.vision_model.encoder.layers.0.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.12.layer_norm2.weight', 'git.vision_model.encoder.layers.1.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.16.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.9.layer_norm1.weight', 'git.vision_model.encoder.layers.0.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.22.layer_norm2.weight', 'git.vision_model.encoder.layers.4.layer_norm2.bias', 'git.vision_model.encoder.layers.22.mlp.fc1.weight', 'git.vision_model.encoder.layers.14.mlp.fc2.bias', 'git.vision_model.encoder.layers.19.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.9.mlp.fc2.weight', 'git.vision_model.encoder.layers.10.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.17.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.18.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.18.layer_norm1.weight', 'git.vision_model.encoder.layers.23.layer_norm2.bias', 'git.vision_model.encoder.layers.5.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.22.mlp.fc1.bias', 'git.vision_model.encoder.layers.6.layer_norm1.weight', 'git.vision_model.encoder.layers.8.mlp.fc1.bias', 'git.vision_model.encoder.layers.9.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.19.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.20.layer_norm1.weight', 'git.vision_model.encoder.layers.16.mlp.fc2.weight', 'git.vision_model.encoder.layers.5.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.7.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.4.layer_norm1.bias', 'git.vision_model.encoder.layers.23.mlp.fc2.weight', 'git.vision_model.encoder.layers.20.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.23.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.21.layer_norm2.bias', 'git.vision_model.encoder.layers.7.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.5.layer_norm1.bias', 'git.vision_model.encoder.layers.18.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.14.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.12.layer_norm2.bias', 'git.vision_model.encoder.layers.1.mlp.fc2.weight', 'git.vision_model.encoder.layers.7.layer_norm1.bias', 'git.vision_model.encoder.layers.11.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.9.layer_norm2.weight', 'git.vision_model.encoder.layers.2.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.18.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.6.layer_norm2.bias', 'git.vision_model.encoder.layers.13.mlp.fc2.bias', 'git.vision_model.post_layernorm.bias', 'git.vision_model.encoder.layers.18.mlp.fc1.bias', 'git.vision_model.encoder.layers.13.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.2.layer_norm1.bias', 'git.vision_model.encoder.layers.15.layer_norm1.weight', 'git.vision_model.encoder.layers.1.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.6.layer_norm1.bias', 'git.vision_model.encoder.layers.14.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.3.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.20.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.15.mlp.fc2.weight', 'git.vision_model.encoder.layers.5.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.21.layer_norm1.weight', 'git.vision_model.encoder.layers.11.layer_norm1.weight', 'git.vision_model.encoder.layers.14.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.19.mlp.fc2.weight', 'git.vision_model.encoder.layers.11.layer_norm2.bias', 'git.vision_model.encoder.layers.16.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.0.mlp.fc2.weight', 'git.vision_model.encoder.layers.4.self_attn.k_proj.weight', 'git.vision_model.embeddings.position_embedding.weight', 'git.vision_model.encoder.layers.21.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.3.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.11.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.17.mlp.fc1.bias', 'git.vision_model.encoder.layers.21.layer_norm2.weight', 'git.vision_model.encoder.layers.7.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.4.mlp.fc1.weight', 'git.vision_model.encoder.layers.10.layer_norm1.weight', 'git.vision_model.encoder.layers.7.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.13.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.22.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.21.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.14.layer_norm1.weight', 'git.vision_model.encoder.layers.3.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.13.mlp.fc1.weight', 'git.vision_model.encoder.layers.4.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.8.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.21.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.12.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.5.layer_norm2.weight', 'git.vision_model.encoder.layers.4.mlp.fc2.weight', 'git.vision_model.encoder.layers.13.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.1.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.22.layer_norm2.bias', 'git.vision_model.encoder.layers.6.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.22.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.15.mlp.fc1.bias', 'git.vision_model.encoder.layers.0.mlp.fc1.weight', 'git.vision_model.encoder.layers.1.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.21.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.22.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.3.mlp.fc2.bias', 'git.vision_model.encoder.layers.3.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.19.mlp.fc2.bias', 'git.vision_model.encoder.layers.3.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.14.mlp.fc1.bias', 'git.vision_model.encoder.layers.19.layer_norm1.bias', 'git.vision_model.encoder.layers.22.mlp.fc2.weight', 'git.vision_model.encoder.layers.13.layer_norm1.weight', 'git.vision_model.encoder.layers.10.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.17.mlp.fc1.weight', 'git.vision_model.encoder.layers.4.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.15.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.6.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.16.mlp.fc1.weight', 'git.vision_model.encoder.layers.0.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.8.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.6.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.15.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.8.mlp.fc2.bias', 'git.vision_model.encoder.layers.21.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.10.mlp.fc2.weight', 'git.vision_model.encoder.layers.1.mlp.fc1.bias', 'git.vision_model.encoder.layers.4.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.9.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.7.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.17.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.14.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.19.mlp.fc1.bias', 'git.vision_model.encoder.layers.13.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.15.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.5.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.8.layer_norm1.weight', 'git.vision_model.encoder.layers.9.layer_norm2.bias', 'git.vision_model.encoder.layers.7.layer_norm2.bias', 'git.vision_model.encoder.layers.9.mlp.fc1.bias', 'git.vision_model.encoder.layers.7.mlp.fc2.weight', 'git.vision_model.encoder.layers.5.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.14.mlp.fc2.weight', 'git.vision_model.encoder.layers.10.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.6.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.18.layer_norm2.weight', 'git.vision_model.encoder.layers.9.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.9.mlp.fc1.weight', 'git.vision_model.post_layernorm.weight', 'git.vision_model.encoder.layers.4.layer_norm2.weight', 'git.vision_model.encoder.layers.3.layer_norm1.weight', 'git.vision_model.encoder.layers.2.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.9.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.6.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.2.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.12.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.4.mlp.fc2.bias', 'git.vision_model.encoder.layers.23.mlp.fc1.weight', 'git.vision_model.encoder.layers.14.mlp.fc1.weight', 'git.vision_model.encoder.layers.3.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.7.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.7.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.17.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.12.layer_norm1.bias', 'git.vision_model.encoder.layers.5.mlp.fc2.weight', 'git.vision_model.encoder.layers.23.layer_norm1.weight', 'git.vision_model.encoder.layers.2.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.7.layer_norm2.weight', 'git.vision_model.encoder.layers.2.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.0.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.3.mlp.fc1.weight', 'git.vision_model.encoder.layers.17.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.20.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.11.mlp.fc1.bias', 'git.vision_model.encoder.layers.18.mlp.fc1.weight', 'git.vision_model.encoder.layers.23.layer_norm1.bias', 'git.vision_model.embeddings.patch_embedding.weight', 'git.vision_model.encoder.layers.1.mlp.fc1.weight', 'git.vision_model.encoder.layers.20.layer_norm2.weight', 'git.vision_model.encoder.layers.2.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.21.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.16.mlp.fc2.bias', 'git.vision_model.encoder.layers.12.mlp.fc2.bias', 'git.vision_model.encoder.layers.21.mlp.fc2.bias', 'git.vision_model.encoder.layers.1.layer_norm1.weight', 'git.vision_model.encoder.layers.16.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.8.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.15.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.16.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.16.layer_norm2.weight', 'git.vision_model.encoder.layers.14.layer_norm2.bias', 'git.vision_model.encoder.layers.9.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.11.layer_norm1.bias', 'git.vision_model.encoder.layers.17.layer_norm1.bias', 'git.vision_model.encoder.layers.5.layer_norm2.bias', 'git.vision_model.encoder.layers.6.mlp.fc2.bias', 'git.vision_model.encoder.layers.1.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.16.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.10.layer_norm2.bias', 'git.vision_model.encoder.layers.18.layer_norm1.bias', 'git.vision_model.encoder.layers.5.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.22.layer_norm1.weight', 'git.vision_model.encoder.layers.8.mlp.fc2.weight', 'git.vision_model.encoder.layers.4.layer_norm1.weight', 'git.vision_model.encoder.layers.5.layer_norm1.weight', 'git.vision_model.encoder.layers.17.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.8.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.23.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.14.layer_norm2.weight', 'git.vision_model.encoder.layers.8.layer_norm1.bias', 'git.vision_model.encoder.layers.0.mlp.fc1.bias', 'git.vision_model.encoder.layers.21.mlp.fc1.bias', 'git.vision_model.encoder.layers.11.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.23.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.1.mlp.fc2.bias', 'git.vision_model.encoder.layers.22.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.11.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.11.mlp.fc2.bias', 'git.vision_model.encoder.layers.10.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.5.mlp.fc1.bias', 'git.vision_model.encoder.layers.9.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.8.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.16.layer_norm1.bias', 'git.vision_model.encoder.layers.19.mlp.fc1.weight', 'git.vision_model.encoder.layers.10.mlp.fc1.weight', 'git.vision_model.encoder.layers.16.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.21.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.10.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.1.layer_norm2.weight', 'git.vision_model.encoder.layers.2.layer_norm2.bias', 'git.vision_model.encoder.layers.18.layer_norm2.bias', 'git.vision_model.encoder.layers.0.mlp.fc2.bias', 'git.vision_model.encoder.layers.12.layer_norm1.weight', 'git.vision_model.encoder.layers.8.layer_norm2.bias', 'git.vision_model.encoder.layers.15.mlp.fc1.weight', 'git.vision_model.encoder.layers.2.layer_norm2.weight', 'git.vision_model.encoder.layers.23.mlp.fc1.bias', 'git.vision_model.encoder.layers.15.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.0.layer_norm1.weight', 'git.vision_model.encoder.layers.11.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.2.layer_norm1.weight', 'git.vision_model.encoder.layers.5.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.13.layer_norm2.weight', 'git.vision_model.encoder.layers.18.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.6.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.1.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.15.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.16.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.3.mlp.fc1.bias', 'git.vision_model.encoder.layers.17.layer_norm2.bias', 'git.vision_model.encoder.layers.17.mlp.fc2.weight', 'git.vision_model.encoder.layers.16.mlp.fc1.bias', 'git.vision_model.encoder.layers.20.self_attn.out_proj.bias', 'git.vision_model.embeddings.class_embedding', 'git.vision_model.encoder.layers.7.mlp.fc1.weight', 'git.vision_model.encoder.layers.6.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.11.mlp.fc1.weight', 'git.vision_model.encoder.layers.3.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.4.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.12.mlp.fc2.weight', 'git.vision_model.encoder.layers.0.layer_norm2.weight', 'git.vision_model.encoder.layers.20.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.10.mlp.fc1.bias', 'git.vision_model.encoder.layers.14.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.22.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.0.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.7.layer_norm1.weight', 'git.vision_model.encoder.layers.8.layer_norm2.weight', 'git.vision_model.encoder.layers.19.layer_norm1.weight', 'git.vision_model.encoder.layers.20.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.8.mlp.fc1.weight', 'git.vision_model.encoder.layers.6.mlp.fc1.weight', 'git.vision_model.encoder.layers.18.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.11.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.21.layer_norm1.bias', 'git.vision_model.encoder.layers.3.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.12.mlp.fc1.weight', 'git.vision_model.encoder.layers.16.layer_norm2.bias', 'git.vision_model.encoder.layers.6.mlp.fc1.bias', 'git.vision_model.encoder.layers.8.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.18.mlp.fc2.bias', 'git.vision_model.encoder.layers.14.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.23.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.10.mlp.fc2.bias', 'git.vision_model.encoder.layers.23.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.18.mlp.fc2.weight', 'git.vision_model.encoder.layers.15.layer_norm2.weight', 'git.vision_model.encoder.layers.2.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.0.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.12.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.4.mlp.fc1.bias', 'git.vision_model.encoder.layers.15.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.18.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.4.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.3.mlp.fc2.weight', 'git.vision_model.encoder.layers.2.mlp.fc2.bias', 'git.vision_model.pre_layrnorm.bias', 'git.vision_model.encoder.layers.17.mlp.fc2.bias', 'git.vision_model.encoder.layers.13.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.20.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.21.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.1.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.17.layer_norm1.weight', 'git.vision_model.encoder.layers.6.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.8.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.22.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.13.mlp.fc1.bias', 'git.vision_model.encoder.layers.19.layer_norm2.bias', 'git.vision_model.encoder.layers.13.layer_norm1.bias', 'git.vision_model.encoder.layers.2.mlp.fc1.bias', 'git.vision_model.encoder.layers.20.mlp.fc1.bias', 'git.vision_model.encoder.layers.11.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.13.mlp.fc2.weight', 'git.vision_model.encoder.layers.9.layer_norm1.bias', 'git.vision_model.encoder.layers.14.layer_norm1.bias', 'git.vision_model.encoder.layers.5.mlp.fc2.bias', 'git.vision_model.encoder.layers.14.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.0.layer_norm2.bias', 'git.vision_model.encoder.layers.16.layer_norm1.weight', 'git.vision_model.encoder.layers.23.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.7.mlp.fc1.bias', 'git.vision_model.encoder.layers.20.mlp.fc1.weight', 'git.vision_model.encoder.layers.13.layer_norm2.bias', 'git.vision_model.encoder.layers.7.self_attn.v_proj.bias', 'git.vision_model.encoder.layers.12.mlp.fc1.bias', 'git.vision_model.encoder.layers.0.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.5.mlp.fc1.weight', 'git.vision_model.encoder.layers.20.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.2.mlp.fc2.weight', 'git.vision_model.encoder.layers.13.self_attn.out_proj.bias', 'git.vision_model.encoder.layers.23.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.3.layer_norm2.bias', 'git.vision_model.encoder.layers.12.self_attn.k_proj.weight', 'git.vision_model.encoder.layers.17.self_attn.q_proj.weight', 'git.vision_model.encoder.layers.1.layer_norm1.bias', 'git.vision_model.encoder.layers.20.mlp.fc2.weight', 'git.vision_model.encoder.layers.11.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.15.mlp.fc2.bias', 'git.vision_model.encoder.layers.0.self_attn.q_proj.bias', 'git.vision_model.encoder.layers.4.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.9.self_attn.k_proj.bias', 'git.vision_model.encoder.layers.19.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.20.mlp.fc2.bias', 'git.vision_model.encoder.layers.3.layer_norm2.weight', 'git.vision_model.encoder.layers.19.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.21.mlp.fc2.weight', 'git.vision_model.encoder.layers.15.layer_norm2.bias', 'git.vision_model.encoder.layers.22.mlp.fc2.bias', 'git.vision_model.encoder.layers.14.self_attn.out_proj.weight', 'git.vision_model.encoder.layers.10.self_attn.v_proj.weight', 'git.vision_model.encoder.layers.6.mlp.fc2.weight', 'git.vision_model.encoder.layers.19.self_attn.k_proj.bias']
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
