Skip to content

Nothing works #369

Open
Open
Nothing works#369
Labels
bugSomething isn't working

Description

@Giddur
Giddur
opened on Oct 8, 2026

Which official extension?

TripoSG

Describe the bug

Error: Traceback (most recent call last):
File "G:\AI Mesh Gen\Modly\resources\api\routers\generation.py", line 317, in _run_generation_impl
gen = get_generator()
^^^^^^^^^^^^^^^
File "G:\AI Mesh Gen\Modly\resources\api\routers\generation.py", line 297, in
get_generator = (lambda: generator_registry.activate_ready_generator(model_id))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "G:\AI Mesh Gen\Modly\resources\api\services\generator_registry.py", line 840, in activate_ready_generator
return self.get_ready_generator(model_id)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "G:\AI Mesh Gen\Modly\resources\api\services\generator_registry.py", line 833, in get_ready_generator
gen.load()
File "G:\AI Mesh Gen\Modly\resources\api\services\extension_process.py", line 288, in load
msg = self._recv(timeout=None) # model load can be arbitrarily slow
^^^^^^^^^^^^^^^^^^^^^^^^
File "G:\AI Mesh Gen\Modly\resources\api\services\extension_process.py", line 265, in _recv
raise RuntimeError(f"[{self.MODEL_ID}] Subprocess died unexpectedly")
RuntimeError: [triposg/generate] Subprocess died unexpectedly

Steps to reproduce

How the hell am I supposed to know?

Modly version

0.3.5

Operating system

10

GPU

2070

Screenshots & logs

No response

Confirmation

  • I have searched existing issues and this has not been reported yet.
  • I can reproduce this on the latest version of Modly.

Activity

  1. Giddur
    added
    bugSomething isn't working
    on Oct 8, 2026

  2. DrCyanide commented on Oct 13, 2026

    @DrCyanide
    DrCyanide
    on Oct 13, 2026

    I'm getting a very similar error on Linux using Hunyuan3D 2 Mini Turbo

    Error: Traceback (most recent call last):
  File "/tmp/.mount_Modly-A54aOO/resources/api/routers/generation.py", line 317, in _run_generation_impl
    gen = get_generator()
          ^^^^^^^^^^^^^^^
  File "/tmp/.mount_Modly-A54aOO/resources/api/routers/generation.py", line 297, in <lambda>
    get_generator = (lambda: generator_registry.activate_ready_generator(model_id)) \
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/tmp/.mount_Modly-A54aOO/resources/api/services/generator_registry.py", line 840, in activate_ready_generator
    return self.get_ready_generator(model_id)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/tmp/.mount_Modly-A54aOO/resources/api/services/generator_registry.py", line 833, in get_ready_generator
    gen.load()
  File "/tmp/.mount_Modly-A54aOO/resources/api/services/extension_process.py", line 292, in load
    raise RuntimeError(msg.get("traceback") or msg.get("message"))
RuntimeError: Traceback (most recent call last):
  File "/tmp/.mount_Modly-A54aOO/resources/api/runner.py", line 239, in main
    gen.load()
  File "/home/username/Documents/Modly/extensions/hunyuan3d-mini-turbo/generator.py", line 72, in load
    pipeline = Hunyuan3DDiTFlowMatchingPipeline.from_pretrained(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/username/Documents/Modly/models/hunyuan3d-mini-turbo/generate/_hy3dgen/hy3dgen/shapegen/pipelines.py", line 225, in from_pretrained
    return cls.from_single_file(
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/username/Documents/Modly/models/hunyuan3d-mini-turbo/generate/_hy3dgen/hy3dgen/shapegen/utils.py", line 83, in wrapper
    result = func(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^
  File "/home/username/Documents/Modly/models/hunyuan3d-mini-turbo/generate/_hy3dgen/hy3dgen/shapegen/pipelines.py", line 181, in from_single_file
    conditioner.load_state_dict(ckpt['conditioner'])
  File "/home/username/Documents/Modly/extensions/hunyuan3d-mini-turbo/venv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 2581, in load_state_dict
    raise RuntimeError(
RuntimeError: Error(s) in loading state_dict for SingleImageEncoder:
	Missing key(s) in state_dict: "main_image_encoder.model.encoder.layer.0.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.0.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.0.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.0.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.0.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.0.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.0.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.0.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.0.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.0.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.0.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.0.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.0.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.0.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.1.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.1.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.1.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.1.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.1.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.1.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.1.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.1.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.1.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.1.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.1.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.1.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.1.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.1.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.2.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.2.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.2.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.2.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.2.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.2.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.2.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.2.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.2.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.2.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.2.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.2.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.2.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.2.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.3.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.3.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.3.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.3.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.3.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.3.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.3.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.3.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.3.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.3.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.3.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.3.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.3.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.3.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.4.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.4.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.4.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.4.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.4.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.4.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.4.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.4.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.4.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.4.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.4.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.4.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.4.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.4.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.5.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.5.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.5.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.5.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.5.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.5.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.5.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.5.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.5.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.5.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.5.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.5.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.5.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.5.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.6.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.6.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.6.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.6.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.6.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.6.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.6.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.6.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.6.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.6.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.6.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.6.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.6.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.6.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.7.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.7.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.7.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.7.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.7.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.7.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.7.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.7.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.7.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.7.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.7.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.7.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.7.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.7.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.8.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.8.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.8.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.8.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.8.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.8.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.8.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.8.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.8.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.8.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.8.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.8.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.8.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.8.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.9.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.9.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.9.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.9.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.9.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.9.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.9.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.9.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.9.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.9.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.9.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.9.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.9.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.9.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.10.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.10.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.10.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.10.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.10.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.10.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.10.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.10.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.10.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.10.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.10.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.10.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.10.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.10.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.11.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.11.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.11.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.11.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.11.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.11.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.11.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.11.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.11.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.11.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.11.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.11.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.11.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.11.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.12.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.12.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.12.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.12.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.12.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.12.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.12.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.12.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.12.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.12.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.12.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.12.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.12.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.12.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.13.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.13.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.13.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.13.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.13.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.13.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.13.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.13.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.13.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.13.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.13.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.13.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.13.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.13.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.14.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.14.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.14.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.14.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.14.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.14.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.14.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.14.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.14.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.14.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.14.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.14.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.14.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.14.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.15.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.15.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.15.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.15.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.15.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.15.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.15.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.15.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.15.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.15.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.15.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.15.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.15.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.15.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.16.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.16.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.16.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.16.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.16.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.16.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.16.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.16.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.16.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.16.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.16.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.16.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.16.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.16.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.17.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.17.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.17.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.17.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.17.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.17.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.17.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.17.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.17.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.17.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.17.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.17.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.17.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.17.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.18.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.18.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.18.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.18.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.18.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.18.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.18.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.18.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.18.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.18.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.18.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.18.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.18.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.18.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.19.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.19.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.19.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.19.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.19.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.19.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.19.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.19.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.19.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.19.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.19.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.19.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.19.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.19.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.20.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.20.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.20.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.20.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.20.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.20.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.20.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.20.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.20.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.20.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.20.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.20.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.20.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.20.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.21.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.21.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.21.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.21.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.21.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.21.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.21.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.21.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.21.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.21.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.21.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.21.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.21.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.21.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.22.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.22.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.22.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.22.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.22.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.22.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.22.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.22.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.22.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.22.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.22.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.22.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.22.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.22.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.23.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.23.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.23.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.23.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.23.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.23.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.23.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.23.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.23.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.23.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.23.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.23.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.23.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.23.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.24.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.24.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.24.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.24.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.24.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.24.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.24.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.24.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.24.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.24.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.24.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.24.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.24.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.24.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.25.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.25.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.25.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.25.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.25.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.25.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.25.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.25.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.25.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.25.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.25.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.25.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.25.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.25.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.26.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.26.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.26.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.26.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.26.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.26.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.26.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.26.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.26.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.26.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.26.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.26.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.26.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.26.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.27.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.27.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.27.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.27.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.27.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.27.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.27.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.27.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.27.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.27.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.27.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.27.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.27.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.27.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.28.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.28.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.28.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.28.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.28.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.28.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.28.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.28.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.28.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.28.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.28.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.28.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.28.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.28.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.29.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.29.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.29.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.29.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.29.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.29.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.29.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.29.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.29.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.29.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.29.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.29.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.29.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.29.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.30.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.30.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.30.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.30.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.30.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.30.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.30.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.30.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.30.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.30.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.30.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.30.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.30.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.30.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.31.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.31.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.31.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.31.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.31.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.31.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.31.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.31.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.31.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.31.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.31.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.31.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.31.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.31.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.32.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.32.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.32.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.32.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.32.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.32.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.32.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.32.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.32.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.32.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.32.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.32.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.32.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.32.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.33.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.33.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.33.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.33.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.33.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.33.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.33.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.33.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.33.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.33.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.33.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.33.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.33.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.33.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.34.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.34.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.34.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.34.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.34.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.34.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.34.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.34.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.34.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.34.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.34.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.34.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.34.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.34.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.35.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.35.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.35.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.35.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.35.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.35.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.35.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.35.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.35.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.35.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.35.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.35.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.35.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.35.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.36.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.36.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.36.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.36.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.36.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.36.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.36.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.36.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.36.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.36.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.36.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.36.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.36.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.36.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.37.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.37.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.37.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.37.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.37.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.37.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.37.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.37.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.37.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.37.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.37.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.37.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.37.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.37.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.38.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.38.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.38.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.38.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.38.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.38.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.38.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.38.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.38.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.38.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.38.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.38.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.38.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.38.mlp.down_proj.bias", "main_image_encoder.model.encoder.layer.39.attention.q_proj.weight", "main_image_encoder.model.encoder.layer.39.attention.q_proj.bias", "main_image_encoder.model.encoder.layer.39.attention.k_proj.weight", "main_image_encoder.model.encoder.layer.39.attention.k_proj.bias", "main_image_encoder.model.encoder.layer.39.attention.v_proj.weight", "main_image_encoder.model.encoder.layer.39.attention.v_proj.bias", "main_image_encoder.model.encoder.layer.39.attention.o_proj.weight", "main_image_encoder.model.encoder.layer.39.attention.o_proj.bias", "main_image_encoder.model.encoder.layer.39.mlp.gate_proj.weight", "main_image_encoder.model.encoder.layer.39.mlp.gate_proj.bias", "main_image_encoder.model.encoder.layer.39.mlp.up_proj.weight", "main_image_encoder.model.encoder.layer.39.mlp.up_proj.bias", "main_image_encoder.model.encoder.layer.39.mlp.down_proj.weight", "main_image_encoder.model.encoder.layer.39.mlp.down_proj.bias". 
	Unexpected key(s) in state_dict: "main_image_encoder.model.encoder.layer.0.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.0.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.0.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.0.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.0.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.0.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.0.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.0.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.0.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.0.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.0.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.0.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.1.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.1.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.1.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.1.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.1.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.1.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.1.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.1.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.1.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.1.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.1.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.1.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.2.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.2.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.2.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.2.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.2.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.2.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.2.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.2.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.2.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.2.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.2.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.2.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.3.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.3.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.3.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.3.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.3.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.3.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.3.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.3.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.3.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.3.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.3.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.3.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.4.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.4.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.4.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.4.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.4.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.4.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.4.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.4.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.4.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.4.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.4.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.4.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.5.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.5.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.5.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.5.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.5.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.5.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.5.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.5.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.5.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.5.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.5.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.5.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.6.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.6.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.6.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.6.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.6.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.6.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.6.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.6.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.6.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.6.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.6.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.6.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.7.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.7.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.7.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.7.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.7.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.7.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.7.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.7.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.7.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.7.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.7.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.7.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.8.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.8.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.8.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.8.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.8.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.8.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.8.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.8.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.8.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.8.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.8.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.8.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.9.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.9.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.9.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.9.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.9.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.9.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.9.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.9.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.9.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.9.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.9.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.9.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.10.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.10.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.10.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.10.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.10.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.10.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.10.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.10.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.10.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.10.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.10.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.10.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.11.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.11.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.11.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.11.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.11.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.11.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.11.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.11.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.11.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.11.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.11.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.11.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.12.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.12.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.12.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.12.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.12.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.12.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.12.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.12.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.12.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.12.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.12.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.12.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.13.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.13.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.13.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.13.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.13.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.13.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.13.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.13.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.13.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.13.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.13.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.13.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.14.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.14.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.14.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.14.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.14.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.14.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.14.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.14.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.14.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.14.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.14.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.14.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.15.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.15.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.15.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.15.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.15.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.15.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.15.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.15.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.15.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.15.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.15.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.15.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.16.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.16.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.16.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.16.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.16.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.16.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.16.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.16.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.16.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.16.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.16.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.16.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.17.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.17.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.17.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.17.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.17.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.17.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.17.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.17.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.17.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.17.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.17.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.17.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.18.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.18.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.18.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.18.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.18.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.18.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.18.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.18.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.18.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.18.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.18.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.18.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.19.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.19.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.19.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.19.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.19.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.19.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.19.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.19.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.19.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.19.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.19.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.19.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.20.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.20.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.20.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.20.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.20.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.20.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.20.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.20.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.20.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.20.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.20.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.20.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.21.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.21.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.21.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.21.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.21.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.21.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.21.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.21.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.21.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.21.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.21.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.21.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.22.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.22.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.22.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.22.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.22.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.22.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.22.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.22.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.22.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.22.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.22.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.22.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.23.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.23.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.23.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.23.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.23.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.23.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.23.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.23.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.23.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.23.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.23.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.23.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.24.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.24.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.24.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.24.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.24.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.24.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.24.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.24.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.24.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.24.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.24.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.24.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.25.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.25.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.25.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.25.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.25.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.25.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.25.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.25.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.25.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.25.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.25.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.25.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.26.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.26.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.26.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.26.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.26.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.26.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.26.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.26.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.26.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.26.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.26.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.26.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.27.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.27.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.27.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.27.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.27.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.27.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.27.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.27.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.27.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.27.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.27.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.27.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.28.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.28.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.28.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.28.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.28.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.28.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.28.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.28.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.28.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.28.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.28.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.28.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.29.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.29.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.29.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.29.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.29.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.29.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.29.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.29.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.29.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.29.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.29.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.29.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.30.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.30.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.30.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.30.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.30.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.30.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.30.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.30.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.30.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.30.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.30.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.30.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.31.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.31.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.31.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.31.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.31.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.31.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.31.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.31.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.31.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.31.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.31.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.31.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.32.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.32.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.32.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.32.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.32.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.32.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.32.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.32.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.32.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.32.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.32.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.32.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.33.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.33.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.33.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.33.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.33.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.33.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.33.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.33.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.33.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.33.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.33.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.33.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.34.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.34.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.34.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.34.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.34.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.34.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.34.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.34.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.34.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.34.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.34.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.34.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.35.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.35.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.35.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.35.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.35.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.35.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.35.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.35.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.35.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.35.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.35.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.35.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.36.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.36.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.36.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.36.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.36.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.36.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.36.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.36.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.36.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.36.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.36.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.36.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.37.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.37.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.37.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.37.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.37.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.37.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.37.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.37.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.37.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.37.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.37.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.37.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.38.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.38.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.38.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.38.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.38.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.38.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.38.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.38.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.38.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.38.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.38.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.38.mlp.weights_out.weight", "main_image_encoder.model.encoder.layer.39.attention.attention.key.bias", "main_image_encoder.model.encoder.layer.39.attention.attention.key.weight", "main_image_encoder.model.encoder.layer.39.attention.attention.query.bias", "main_image_encoder.model.encoder.layer.39.attention.attention.query.weight", "main_image_encoder.model.encoder.layer.39.attention.attention.value.bias", "main_image_encoder.model.encoder.layer.39.attention.attention.value.weight", "main_image_encoder.model.encoder.layer.39.attention.output.dense.bias", "main_image_encoder.model.encoder.layer.39.attention.output.dense.weight", "main_image_encoder.model.encoder.layer.39.mlp.weights_in.bias", "main_image_encoder.model.encoder.layer.39.mlp.weights_in.weight", "main_image_encoder.model.encoder.layer.39.mlp.weights_out.bias", "main_image_encoder.model.encoder.layer.39.mlp.weights_out.weight".

    That last part, the RuntimeError, makes it seem like Modly made some change that broke Extensions and then didn't update the extensions

  3. krol0278-source commented on Oct 13, 2026

    @krol0278-source
    krol0278-source
    on Oct 13, 2026

    Identical error

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't working

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions