gua: stage-B LoRA SFT of Qwen3.5-4B for macOS computer-use decisions
Browse files- .gitattributes +1 -0
- LICENSE +201 -0
- README.md +192 -0
- adapter/adapter_config.json +259 -0
- adapter/adapters.safetensors +3 -0
- chat_template.jinja +154 -0
- config.json +104 -0
- inference/cua_decide.py +318 -0
- merge_check.json +5 -0
- model-00001-of-00002.safetensors +3 -0
- model-00002-of-00002.safetensors +3 -0
- model.safetensors.index.json +730 -0
- preprocessor_config.json +21 -0
- processor_config.json +63 -0
- temperatures.json +69 -0
- tokenizer.json +3 -0
- tokenizer_config.json +32 -0
- video_preprocessor_config.json +21 -0
- vocab.json +0 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
LICENSE
ADDED
|
@@ -0,0 +1,201 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Apache License
|
| 2 |
+
Version 2.0, January 2004
|
| 3 |
+
http://www.apache.org/licenses/
|
| 4 |
+
|
| 5 |
+
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
|
| 6 |
+
|
| 7 |
+
1. Definitions.
|
| 8 |
+
|
| 9 |
+
"License" shall mean the terms and conditions for use, reproduction,
|
| 10 |
+
and distribution as defined by Sections 1 through 9 of this document.
|
| 11 |
+
|
| 12 |
+
"Licensor" shall mean the copyright owner or entity authorized by
|
| 13 |
+
the copyright owner that is granting the License.
|
| 14 |
+
|
| 15 |
+
"Legal Entity" shall mean the union of the acting entity and all
|
| 16 |
+
other entities that control, are controlled by, or are under common
|
| 17 |
+
control with that entity. For the purposes of this definition,
|
| 18 |
+
"control" means (i) the power, direct or indirect, to cause the
|
| 19 |
+
direction or management of such entity, whether by contract or
|
| 20 |
+
otherwise, or (ii) ownership of fifty percent (50%) or more of the
|
| 21 |
+
outstanding shares, or (iii) beneficial ownership of such entity.
|
| 22 |
+
|
| 23 |
+
"You" (or "Your") shall mean an individual or Legal Entity
|
| 24 |
+
exercising permissions granted by this License.
|
| 25 |
+
|
| 26 |
+
"Source" form shall mean the preferred form for making modifications,
|
| 27 |
+
including but not limited to software source code, documentation
|
| 28 |
+
source, and configuration files.
|
| 29 |
+
|
| 30 |
+
"Object" form shall mean any form resulting from mechanical
|
| 31 |
+
transformation or translation of a Source form, including but
|
| 32 |
+
not limited to compiled object code, generated documentation,
|
| 33 |
+
and conversions to other media types.
|
| 34 |
+
|
| 35 |
+
"Work" shall mean the work of authorship, whether in Source or
|
| 36 |
+
Object form, made available under the License, as indicated by a
|
| 37 |
+
copyright notice that is included in or attached to the work
|
| 38 |
+
(an example is provided in the Appendix below).
|
| 39 |
+
|
| 40 |
+
"Derivative Works" shall mean any work, whether in Source or Object
|
| 41 |
+
form, that is based on (or derived from) the Work and for which the
|
| 42 |
+
editorial revisions, annotations, elaborations, or other modifications
|
| 43 |
+
represent, as a whole, an original work of authorship. For the purposes
|
| 44 |
+
of this License, Derivative Works shall not include works that remain
|
| 45 |
+
separable from, or merely link (or bind by name) to the interfaces of,
|
| 46 |
+
the Work and Derivative Works thereof.
|
| 47 |
+
|
| 48 |
+
"Contribution" shall mean any work of authorship, including
|
| 49 |
+
the original version of the Work and any modifications or additions
|
| 50 |
+
to that Work or Derivative Works thereof, that is intentionally
|
| 51 |
+
submitted to Licensor for inclusion in the Work by the copyright owner
|
| 52 |
+
or by an individual or Legal Entity authorized to submit on behalf of
|
| 53 |
+
the copyright owner. For the purposes of this definition, "submitted"
|
| 54 |
+
means any form of electronic, verbal, or written communication sent
|
| 55 |
+
to the Licensor or its representatives, including but not limited to
|
| 56 |
+
communication on electronic mailing lists, source code control systems,
|
| 57 |
+
and issue tracking systems that are managed by, or on behalf of, the
|
| 58 |
+
Licensor for the purpose of discussing and improving the Work, but
|
| 59 |
+
excluding communication that is conspicuously marked or otherwise
|
| 60 |
+
designated in writing by the copyright owner as "Not a Contribution."
|
| 61 |
+
|
| 62 |
+
"Contributor" shall mean Licensor and any individual or Legal Entity
|
| 63 |
+
on behalf of whom a Contribution has been received by Licensor and
|
| 64 |
+
subsequently incorporated within the Work.
|
| 65 |
+
|
| 66 |
+
2. Grant of Copyright License. Subject to the terms and conditions of
|
| 67 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 68 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 69 |
+
copyright license to reproduce, prepare Derivative Works of,
|
| 70 |
+
publicly display, publicly perform, sublicense, and distribute the
|
| 71 |
+
Work and such Derivative Works in Source or Object form.
|
| 72 |
+
|
| 73 |
+
3. Grant of Patent License. Subject to the terms and conditions of
|
| 74 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 75 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 76 |
+
(except as stated in this section) patent license to make, have made,
|
| 77 |
+
use, offer to sell, sell, import, and otherwise transfer the Work,
|
| 78 |
+
where such license applies only to those patent claims licensable
|
| 79 |
+
by such Contributor that are necessarily infringed by their
|
| 80 |
+
Contribution(s) alone or by combination of their Contribution(s)
|
| 81 |
+
with the Work to which such Contribution(s) was submitted. If You
|
| 82 |
+
institute patent litigation against any entity (including a
|
| 83 |
+
cross-claim or counterclaim in a lawsuit) alleging that the Work
|
| 84 |
+
or a Contribution incorporated within the Work constitutes direct
|
| 85 |
+
or contributory patent infringement, then any patent licenses
|
| 86 |
+
granted to You under this License for that Work shall terminate
|
| 87 |
+
as of the date such litigation is filed.
|
| 88 |
+
|
| 89 |
+
4. Redistribution. You may reproduce and distribute copies of the
|
| 90 |
+
Work or Derivative Works thereof in any medium, with or without
|
| 91 |
+
modifications, and in Source or Object form, provided that You
|
| 92 |
+
meet the following conditions:
|
| 93 |
+
|
| 94 |
+
(a) You must give any other recipients of the Work or
|
| 95 |
+
Derivative Works a copy of this License; and
|
| 96 |
+
|
| 97 |
+
(b) You must cause any modified files to carry prominent notices
|
| 98 |
+
stating that You changed the files; and
|
| 99 |
+
|
| 100 |
+
(c) You must retain, in the Source form of any Derivative Works
|
| 101 |
+
that You distribute, all copyright, patent, trademark, and
|
| 102 |
+
attribution notices from the Source form of the Work,
|
| 103 |
+
excluding those notices that do not pertain to any part of
|
| 104 |
+
the Derivative Works; and
|
| 105 |
+
|
| 106 |
+
(d) If the Work includes a "NOTICE" text file as part of its
|
| 107 |
+
distribution, then any Derivative Works that You distribute must
|
| 108 |
+
include a readable copy of the attribution notices contained
|
| 109 |
+
within such NOTICE file, excluding those notices that do not
|
| 110 |
+
pertain to any part of the Derivative Works, in at least one
|
| 111 |
+
of the following places: within a NOTICE text file distributed
|
| 112 |
+
as part of the Derivative Works; within the Source form or
|
| 113 |
+
documentation, if provided along with the Derivative Works; or,
|
| 114 |
+
within a display generated by the Derivative Works, if and
|
| 115 |
+
wherever such third-party notices normally appear. The contents
|
| 116 |
+
of the NOTICE file are for informational purposes only and
|
| 117 |
+
do not modify the License. You may add Your own attribution
|
| 118 |
+
notices within Derivative Works that You distribute, alongside
|
| 119 |
+
or as an addendum to the NOTICE text from the Work, provided
|
| 120 |
+
that such additional attribution notices cannot be construed
|
| 121 |
+
as modifying the License.
|
| 122 |
+
|
| 123 |
+
You may add Your own copyright statement to Your modifications and
|
| 124 |
+
may provide additional or different license terms and conditions
|
| 125 |
+
for use, reproduction, or distribution of Your modifications, or
|
| 126 |
+
for any such Derivative Works as a whole, provided Your use,
|
| 127 |
+
reproduction, and distribution of the Work otherwise complies with
|
| 128 |
+
the conditions stated in this License.
|
| 129 |
+
|
| 130 |
+
5. Submission of Contributions. Unless You explicitly state otherwise,
|
| 131 |
+
any Contribution intentionally submitted for inclusion in the Work
|
| 132 |
+
by You to the Licensor shall be under the terms and conditions of
|
| 133 |
+
this License, without any additional terms or conditions.
|
| 134 |
+
Notwithstanding the above, nothing herein shall supersede or modify
|
| 135 |
+
the terms of any separate license agreement you may have executed
|
| 136 |
+
with Licensor regarding such Contributions.
|
| 137 |
+
|
| 138 |
+
6. Trademarks. This License does not grant permission to use the trade
|
| 139 |
+
names, trademarks, service marks, or product names of the Licensor,
|
| 140 |
+
except as required for reasonable and customary use in describing the
|
| 141 |
+
origin of the Work and reproducing the content of the NOTICE file.
|
| 142 |
+
|
| 143 |
+
7. Disclaimer of Warranty. Unless required by applicable law or
|
| 144 |
+
agreed to in writing, Licensor provides the Work (and each
|
| 145 |
+
Contributor provides its Contributions) on an "AS IS" BASIS,
|
| 146 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
| 147 |
+
implied, including, without limitation, any warranties or conditions
|
| 148 |
+
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
|
| 149 |
+
PARTICULAR PURPOSE. You are solely responsible for determining the
|
| 150 |
+
appropriateness of using or redistributing the Work and assume any
|
| 151 |
+
risks associated with Your exercise of permissions under this License.
|
| 152 |
+
|
| 153 |
+
8. Limitation of Liability. In no event and under no legal theory,
|
| 154 |
+
whether in tort (including negligence), contract, or otherwise,
|
| 155 |
+
unless required by applicable law (such as deliberate and grossly
|
| 156 |
+
negligent acts) or agreed to in writing, shall any Contributor be
|
| 157 |
+
liable to You for damages, including any direct, indirect, special,
|
| 158 |
+
incidental, or consequential damages of any character arising as a
|
| 159 |
+
result of this License or out of the use or inability to use the
|
| 160 |
+
Work (including but not limited to damages for loss of goodwill,
|
| 161 |
+
work stoppage, computer failure or malfunction, or any and all
|
| 162 |
+
other commercial damages or losses), even if such Contributor
|
| 163 |
+
has been advised of the possibility of such damages.
|
| 164 |
+
|
| 165 |
+
9. Accepting Warranty or Additional Liability. While redistributing
|
| 166 |
+
the Work or Derivative Works thereof, You may choose to offer,
|
| 167 |
+
and charge a fee for, acceptance of support, warranty, indemnity,
|
| 168 |
+
or other liability obligations and/or rights consistent with this
|
| 169 |
+
License. However, in accepting such obligations, You may act only
|
| 170 |
+
on Your own behalf and on Your sole responsibility, not on behalf
|
| 171 |
+
of any other Contributor, and only if You agree to indemnify,
|
| 172 |
+
defend, and hold each Contributor harmless for any liability
|
| 173 |
+
incurred by, or claims asserted against, such Contributor by reason
|
| 174 |
+
of your accepting any such warranty or additional liability.
|
| 175 |
+
|
| 176 |
+
END OF TERMS AND CONDITIONS
|
| 177 |
+
|
| 178 |
+
APPENDIX: How to apply the Apache License to your work.
|
| 179 |
+
|
| 180 |
+
To apply the Apache License to your work, attach the following
|
| 181 |
+
boilerplate notice, with the fields enclosed by brackets "[]"
|
| 182 |
+
replaced with your own identifying information. (Don't include
|
| 183 |
+
the brackets!) The text should be enclosed in the appropriate
|
| 184 |
+
comment syntax for the file format. We also recommend that a
|
| 185 |
+
file or class name and description of purpose be included on the
|
| 186 |
+
same "printed page" as the copyright notice for easier
|
| 187 |
+
identification within third-party archives.
|
| 188 |
+
|
| 189 |
+
Copyright [yyyy] [name of copyright owner]
|
| 190 |
+
|
| 191 |
+
Licensed under the Apache License, Version 2.0 (the "License");
|
| 192 |
+
you may not use this file except in compliance with the License.
|
| 193 |
+
You may obtain a copy of the License at
|
| 194 |
+
|
| 195 |
+
http://www.apache.org/licenses/LICENSE-2.0
|
| 196 |
+
|
| 197 |
+
Unless required by applicable law or agreed to in writing, software
|
| 198 |
+
distributed under the License is distributed on an "AS IS" BASIS,
|
| 199 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
| 200 |
+
See the License for the specific language governing permissions and
|
| 201 |
+
limitations under the License.
|
README.md
ADDED
|
@@ -0,0 +1,192 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Qwen/Qwen3.5-4B
|
| 3 |
+
library_name: mlx
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
pipeline_tag: image-text-to-text
|
| 6 |
+
tags:
|
| 7 |
+
- mlx
|
| 8 |
+
- qwen3.5
|
| 9 |
+
- lora
|
| 10 |
+
- computer-use
|
| 11 |
+
- gui-agent
|
| 12 |
+
- macos
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
# gua
|
| 16 |
+
|
| 17 |
+
**gua** is a computer-use decision model for macOS. Given a screenshot, the task and the last few
|
| 18 |
+
actions, it chooses **what kind of action comes next** and **which on-screen element it
|
| 19 |
+
acts on**, and can also **ground an element from a description**. It answers as typed
|
| 20 |
+
multiple-choice decisions and returns a probability for every option, so an agent can act
|
| 21 |
+
on confident answers and defer on the rest.
|
| 22 |
+
|
| 23 |
+
It is [Qwen/Qwen3.5-4B](https://huggingface.co/Qwen/Qwen3.5-4B) fine-tuned (supervised
|
| 24 |
+
fine-tuning with LoRA, rank 8 on every linear layer of the language model) on GUI
|
| 25 |
+
trajectories extracted from screen-recorded software tutorials. This repository holds the
|
| 26 |
+
LoRA merged into the base weights (MLX, bf16) and, in `adapter/`, the LoRA itself.
|
| 27 |
+
|
| 28 |
+
> Research model. The training data was derived from public YouTube tutorials and is not
|
| 29 |
+
> released. See [Training data](#training-data) and [Limitations](#limitations).
|
| 30 |
+
|
| 31 |
+
## How it answers
|
| 32 |
+
|
| 33 |
+
The model does not generate free text. Each step is asked as up to three questions, all
|
| 34 |
+
sharing one prompt prefix (system prompt, screenshot, task, past actions, numbered element
|
| 35 |
+
list), and the probability of every option is read from the model's next-token
|
| 36 |
+
distribution:
|
| 37 |
+
|
| 38 |
+
| Question | Options | Use |
|
| 39 |
+
|---|---|---|
|
| 40 |
+
| `action` | click, double-click, right-click, drag, type, press, shortcut, scroll, hover, wait, finish | the kind of the next action |
|
| 41 |
+
| `next_target` | the numbered elements on screen | where the next action goes |
|
| 42 |
+
| `grounding` | the numbered elements on screen | which element matches a description you give |
|
| 43 |
+
|
| 44 |
+
Elements are **candidates found on the screenshot**: text lines from the macOS Vision
|
| 45 |
+
framework, and, during training and evaluation, icons from the
|
| 46 |
+
[OmniParser v2](https://huggingface.co/microsoft/OmniParser-v2.0) icon detector. The model
|
| 47 |
+
chooses among candidates; it cannot point at something no candidate covers. When running
|
| 48 |
+
on a live Mac, the accessibility tree is a better candidate source than detection.
|
| 49 |
+
|
| 50 |
+
The prompt format is fixed by training. Use `inference/cua_decide.py`, which reproduces it
|
| 51 |
+
exactly (checked to give identical option probabilities to the evaluation code).
|
| 52 |
+
|
| 53 |
+
## Usage
|
| 54 |
+
|
| 55 |
+
Apple Silicon with [mlx-vlm](https://github.com/Blaizzy/mlx-vlm) 0.7:
|
| 56 |
+
|
| 57 |
+
```bash
|
| 58 |
+
pip install "mlx-vlm>=0.7,<0.8" pyobjc-framework-vision pyobjc-framework-quartz
|
| 59 |
+
python inference/cua_decide.py --model <user>/gua \
|
| 60 |
+
--image screen.png --task "Turn on Dark Mode" \
|
| 61 |
+
--past "Click the Apple menu" --past "Click System Settings" \
|
| 62 |
+
--describe "the Appearance item in the sidebar" \
|
| 63 |
+
--temperatures temperatures.json
|
| 64 |
+
```
|
| 65 |
+
|
| 66 |
+
Each question prints its choice and calibrated confidence, for example:
|
| 67 |
+
|
| 68 |
+
```json
|
| 69 |
+
{"question": "action", "choice": "click", "confidence": 0.94, "action": "click"}
|
| 70 |
+
{"question": "next_target", "choice": "15", "confidence": 0.60, "element": {"text": "...", "centre": [0.033, 0.136], "box": [...]}}
|
| 71 |
+
```
|
| 72 |
+
|
| 73 |
+
To use the LoRA instead of the merged weights, pass the base model and the adapter:
|
| 74 |
+
`--model mlx-community/Qwen3.5-4B-MLX-bf16 --adapter adapter`.
|
| 75 |
+
|
| 76 |
+
OmniParser's icon detector improves candidate recall but its weights are AGPL-3.0 and are
|
| 77 |
+
not included; pass `--icon-weights icon_detect/model.pt` if you download them yourself.
|
| 78 |
+
Without it, only text candidates are offered, which lowers recall on icon-only controls.
|
| 79 |
+
|
| 80 |
+
## Training
|
| 81 |
+
|
| 82 |
+
| | |
|
| 83 |
+
|---|---|
|
| 84 |
+
| Method | Supervised fine-tuning with LoRA: rank 8, alpha 16, on every linear layer of the language model (`q/k/v/o_proj`, linear-attention `in_proj_*`/`out_proj`, MLP `gate/up/down_proj`); vision tower frozen |
|
| 85 |
+
| Objective | Log loss of the right option label (and end-of-turn token) only, never of the prompt |
|
| 86 |
+
| Data | The first 8,000 of 20,623 shuffled training questions of our dataset *macos-cua v2* (see below) |
|
| 87 |
+
| Optimiser | Adam, learning rate 5e-5, batch 1, gradient clipping 1.0, one pass |
|
| 88 |
+
| Hardware | One Apple Silicon Mac (MLX), gradient checkpointing, about 48 h, peak 161 GB |
|
| 89 |
+
|
| 90 |
+
The recipe was chosen on a held-out dev split, never on the eval split: on the same 2,000
|
| 91 |
+
samples, LoRA on every linear layer matched rank 64 and edged out attention-only LoRA, and
|
| 92 |
+
a full-parameter run of the same model on earlier data did not beat LoRA.
|
| 93 |
+
|
| 94 |
+
Dev score during training (mean accuracy of the three questions on 200 fixed dev steps):
|
| 95 |
+
|
| 96 |
+
| Training samples | 0 (base) | 2,000 | 4,000 | 6,000 | 7,000 | 8,000 |
|
| 97 |
+
|---|---|---|---|---|---|---|
|
| 98 |
+
| Dev score | 0.322 | 0.429 | 0.421 | 0.441 | 0.445 | 0.447 |
|
| 99 |
+
|
| 100 |
+
## Training data
|
| 101 |
+
|
| 102 |
+
Trajectories were extracted automatically from screen-recorded software tutorials on
|
| 103 |
+
YouTube: videos were searched and screened (recording quality, platform), segmented and
|
| 104 |
+
annotated into step-by-step actions by vision-language models, each click was located on
|
| 105 |
+
the frame, and before/after screenshots were cut. The **training split** of *macos-cua v2*:
|
| 106 |
+
|
| 107 |
+
| Topic | Videos | Steps |
|
| 108 |
+
|---|---|---|
|
| 109 |
+
| Godot 4 (game engine) | 79 | 6,271 |
|
| 110 |
+
| Blender (3D) | 39 | 3,764 |
|
| 111 |
+
| macOS video editing (Final Cut Pro, Premiere Pro, CapCut, iMovie) | 7 | 640 |
|
| 112 |
+
| macOS system and productivity apps | 6 | 420 |
|
| 113 |
+
|
| 114 |
+
Splits are by video (dev 14 videos, eval 30 videos, 20 applications appear only in eval).
|
| 115 |
+
Godot typing steps were left out of training: an audit found their labels mostly wrong.
|
| 116 |
+
Many Blender and Godot tutorials were recorded on Windows or full screen; steps that
|
| 117 |
+
would be wrong on a Mac (Windows shortcuts, taskbar, native dialogs) were removed.
|
| 118 |
+
|
| 119 |
+
**Label quality.** Labels are model-generated, not human-verified. A stronger model audited
|
| 120 |
+
a random sample per topic:
|
| 121 |
+
|
| 122 |
+
| Topic | Audited steps | Element located correctly | Action correct |
|
| 123 |
+
|---|---|---|---|
|
| 124 |
+
| macOS system and productivity | 200 | 91.5% | 75.5% |
|
| 125 |
+
| macOS video editing | 395 | 92.4% | 78.2% |
|
| 126 |
+
| Blender | 150 | 89.3% | 70.7% |
|
| 127 |
+
| Godot | 270 | 87.0% | 54.8% |
|
| 128 |
+
|
| 129 |
+
The dataset itself is not released.
|
| 130 |
+
|
| 131 |
+
## Evaluation
|
| 132 |
+
|
| 133 |
+
On the eval split of *macos-cua v2* (3,442 steps from 30 videos never seen in training;
|
| 134 |
+
2,079 steps with an element target), against the untrained base with the same prompt and
|
| 135 |
+
candidates:
|
| 136 |
+
|
| 137 |
+
| Question | Base | This model | Difference (95% CI, bootstrap paired by trajectory) |
|
| 138 |
+
|---|---|---|---|
|
| 139 |
+
| `action` accuracy | 0.481 | 0.525 | +0.043 (+0.005, +0.089) |
|
| 140 |
+
| `grounding` accuracy | 0.443 | 0.603 | +0.160 (+0.132, +0.186) |
|
| 141 |
+
| `next_target` accuracy | 0.155 | 0.261 | +0.106 (+0.085, +0.127) |
|
| 142 |
+
| `action` and `next_target` both right | 0.117 | 0.183 | |
|
| 143 |
+
|
| 144 |
+
On **applications never seen in training** (905 steps from software that appears only in
|
| 145 |
+
the eval split) the gains hold: action 0.439 → 0.550, grounding 0.501 → 0.650, next target
|
| 146 |
+
0.221 → 0.313. A candidate covers the target in 83.4% of the element steps, which bounds
|
| 147 |
+
the element questions.
|
| 148 |
+
|
| 149 |
+
Excluding the 527 Godot typing steps, whose labels an audit found mostly wrong, action
|
| 150 |
+
accuracy is 0.505 → 0.617 (+0.111). Because those steps were left out of training, the
|
| 151 |
+
model rarely answers `type` (22 times against 588 typing steps in eval): expect low recall
|
| 152 |
+
for typing.
|
| 153 |
+
|
| 154 |
+
**Selective accuracy.** Acting only on the most confident half of the answers, grounding
|
| 155 |
+
accuracy is 0.822 and action accuracy 0.636.
|
| 156 |
+
|
| 157 |
+
**Calibration.** `temperatures.json` holds one temperature per question, fitted on the dev
|
| 158 |
+
split by log loss and applied as `p^(1/T)` renormalised (`inference/cua_decide.py
|
| 159 |
+
--temperatures`). Expected calibration error on eval, which the fit never saw:
|
| 160 |
+
|
| 161 |
+
| Question | Temperature | ECE before | ECE after |
|
| 162 |
+
|---|---|---|---|
|
| 163 |
+
| `action` | 2.05 | 0.259 | 0.047 |
|
| 164 |
+
| `grounding` | 1.20 | 0.145 | 0.107 |
|
| 165 |
+
| `next_target` | 0.90 | 0.064 | 0.085 |
|
| 166 |
+
|
| 167 |
+
The action head is strongly overconfident without scaling; the next-target head is
|
| 168 |
+
already close to calibrated, and its fitted temperature slightly lowers log loss on eval
|
| 169 |
+
while slightly raising ECE.
|
| 170 |
+
|
| 171 |
+
## Limitations
|
| 172 |
+
|
| 173 |
+
- **macOS screenshots only**, English UI and tasks; tutorials over-represent Godot and
|
| 174 |
+
Blender.
|
| 175 |
+
- **Candidate-bound.** The model picks among candidates; if no candidate covers the target
|
| 176 |
+
(16.6% of the element steps in our eval), the element questions have no right answer.
|
| 177 |
+
- **Noisy labels**, as audited above; action-type labels are the weakest.
|
| 178 |
+
- **Not an agent by itself.** It decides the action kind and target; the text to type, the
|
| 179 |
+
keys of a shortcut and drag end points are not predicted.
|
| 180 |
+
- **Merged weights round to bf16.** Against base plus adapter, option probabilities differ
|
| 181 |
+
by at most 0.04 in total variation on our check; use `adapter/` for exact reproduction.
|
| 182 |
+
- Confidence is only meaningful after temperature scaling, and only for the question types
|
| 183 |
+
and candidate sources it was fitted with.
|
| 184 |
+
- Rarely predicts `type` (see Evaluation).
|
| 185 |
+
|
| 186 |
+
## Licence and provenance
|
| 187 |
+
|
| 188 |
+
The weights are a derivative of Qwen3.5-4B (Apache-2.0) and are released under Apache-2.0
|
| 189 |
+
(`LICENSE`); the modification is the LoRA fine-tuning described above. The training data
|
| 190 |
+
was derived from public YouTube videos for research; no video frames are distributed here.
|
| 191 |
+
`inference/cua_decide.py` is released under the same licence. OmniParser v2 (optional at
|
| 192 |
+
inference) is licensed separately by its authors.
|
adapter/adapter_config.json
ADDED
|
@@ -0,0 +1,259 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"fine_tune_type": "lora",
|
| 3 |
+
"num_layers": -1,
|
| 4 |
+
"lora_parameters": {
|
| 5 |
+
"rank": 8,
|
| 6 |
+
"dropout": 0,
|
| 7 |
+
"scale": 2.0,
|
| 8 |
+
"keys": [
|
| 9 |
+
"language_model.model.layers.31.mlp.up_proj",
|
| 10 |
+
"language_model.model.layers.31.mlp.down_proj",
|
| 11 |
+
"language_model.model.layers.31.mlp.gate_proj",
|
| 12 |
+
"language_model.model.layers.31.self_attn.o_proj",
|
| 13 |
+
"language_model.model.layers.31.self_attn.v_proj",
|
| 14 |
+
"language_model.model.layers.31.self_attn.k_proj",
|
| 15 |
+
"language_model.model.layers.31.self_attn.q_proj",
|
| 16 |
+
"language_model.model.layers.30.mlp.up_proj",
|
| 17 |
+
"language_model.model.layers.30.mlp.down_proj",
|
| 18 |
+
"language_model.model.layers.30.mlp.gate_proj",
|
| 19 |
+
"language_model.model.layers.30.linear_attn.out_proj",
|
| 20 |
+
"language_model.model.layers.30.linear_attn.in_proj_a",
|
| 21 |
+
"language_model.model.layers.30.linear_attn.in_proj_b",
|
| 22 |
+
"language_model.model.layers.30.linear_attn.in_proj_z",
|
| 23 |
+
"language_model.model.layers.30.linear_attn.in_proj_qkv",
|
| 24 |
+
"language_model.model.layers.29.mlp.up_proj",
|
| 25 |
+
"language_model.model.layers.29.mlp.down_proj",
|
| 26 |
+
"language_model.model.layers.29.mlp.gate_proj",
|
| 27 |
+
"language_model.model.layers.29.linear_attn.out_proj",
|
| 28 |
+
"language_model.model.layers.29.linear_attn.in_proj_a",
|
| 29 |
+
"language_model.model.layers.29.linear_attn.in_proj_b",
|
| 30 |
+
"language_model.model.layers.29.linear_attn.in_proj_z",
|
| 31 |
+
"language_model.model.layers.29.linear_attn.in_proj_qkv",
|
| 32 |
+
"language_model.model.layers.28.mlp.up_proj",
|
| 33 |
+
"language_model.model.layers.28.mlp.down_proj",
|
| 34 |
+
"language_model.model.layers.28.mlp.gate_proj",
|
| 35 |
+
"language_model.model.layers.28.linear_attn.out_proj",
|
| 36 |
+
"language_model.model.layers.28.linear_attn.in_proj_a",
|
| 37 |
+
"language_model.model.layers.28.linear_attn.in_proj_b",
|
| 38 |
+
"language_model.model.layers.28.linear_attn.in_proj_z",
|
| 39 |
+
"language_model.model.layers.28.linear_attn.in_proj_qkv",
|
| 40 |
+
"language_model.model.layers.27.mlp.up_proj",
|
| 41 |
+
"language_model.model.layers.27.mlp.down_proj",
|
| 42 |
+
"language_model.model.layers.27.mlp.gate_proj",
|
| 43 |
+
"language_model.model.layers.27.self_attn.o_proj",
|
| 44 |
+
"language_model.model.layers.27.self_attn.v_proj",
|
| 45 |
+
"language_model.model.layers.27.self_attn.k_proj",
|
| 46 |
+
"language_model.model.layers.27.self_attn.q_proj",
|
| 47 |
+
"language_model.model.layers.26.mlp.up_proj",
|
| 48 |
+
"language_model.model.layers.26.mlp.down_proj",
|
| 49 |
+
"language_model.model.layers.26.mlp.gate_proj",
|
| 50 |
+
"language_model.model.layers.26.linear_attn.out_proj",
|
| 51 |
+
"language_model.model.layers.26.linear_attn.in_proj_a",
|
| 52 |
+
"language_model.model.layers.26.linear_attn.in_proj_b",
|
| 53 |
+
"language_model.model.layers.26.linear_attn.in_proj_z",
|
| 54 |
+
"language_model.model.layers.26.linear_attn.in_proj_qkv",
|
| 55 |
+
"language_model.model.layers.25.mlp.up_proj",
|
| 56 |
+
"language_model.model.layers.25.mlp.down_proj",
|
| 57 |
+
"language_model.model.layers.25.mlp.gate_proj",
|
| 58 |
+
"language_model.model.layers.25.linear_attn.out_proj",
|
| 59 |
+
"language_model.model.layers.25.linear_attn.in_proj_a",
|
| 60 |
+
"language_model.model.layers.25.linear_attn.in_proj_b",
|
| 61 |
+
"language_model.model.layers.25.linear_attn.in_proj_z",
|
| 62 |
+
"language_model.model.layers.25.linear_attn.in_proj_qkv",
|
| 63 |
+
"language_model.model.layers.24.mlp.up_proj",
|
| 64 |
+
"language_model.model.layers.24.mlp.down_proj",
|
| 65 |
+
"language_model.model.layers.24.mlp.gate_proj",
|
| 66 |
+
"language_model.model.layers.24.linear_attn.out_proj",
|
| 67 |
+
"language_model.model.layers.24.linear_attn.in_proj_a",
|
| 68 |
+
"language_model.model.layers.24.linear_attn.in_proj_b",
|
| 69 |
+
"language_model.model.layers.24.linear_attn.in_proj_z",
|
| 70 |
+
"language_model.model.layers.24.linear_attn.in_proj_qkv",
|
| 71 |
+
"language_model.model.layers.23.mlp.up_proj",
|
| 72 |
+
"language_model.model.layers.23.mlp.down_proj",
|
| 73 |
+
"language_model.model.layers.23.mlp.gate_proj",
|
| 74 |
+
"language_model.model.layers.23.self_attn.o_proj",
|
| 75 |
+
"language_model.model.layers.23.self_attn.v_proj",
|
| 76 |
+
"language_model.model.layers.23.self_attn.k_proj",
|
| 77 |
+
"language_model.model.layers.23.self_attn.q_proj",
|
| 78 |
+
"language_model.model.layers.22.mlp.up_proj",
|
| 79 |
+
"language_model.model.layers.22.mlp.down_proj",
|
| 80 |
+
"language_model.model.layers.22.mlp.gate_proj",
|
| 81 |
+
"language_model.model.layers.22.linear_attn.out_proj",
|
| 82 |
+
"language_model.model.layers.22.linear_attn.in_proj_a",
|
| 83 |
+
"language_model.model.layers.22.linear_attn.in_proj_b",
|
| 84 |
+
"language_model.model.layers.22.linear_attn.in_proj_z",
|
| 85 |
+
"language_model.model.layers.22.linear_attn.in_proj_qkv",
|
| 86 |
+
"language_model.model.layers.21.mlp.up_proj",
|
| 87 |
+
"language_model.model.layers.21.mlp.down_proj",
|
| 88 |
+
"language_model.model.layers.21.mlp.gate_proj",
|
| 89 |
+
"language_model.model.layers.21.linear_attn.out_proj",
|
| 90 |
+
"language_model.model.layers.21.linear_attn.in_proj_a",
|
| 91 |
+
"language_model.model.layers.21.linear_attn.in_proj_b",
|
| 92 |
+
"language_model.model.layers.21.linear_attn.in_proj_z",
|
| 93 |
+
"language_model.model.layers.21.linear_attn.in_proj_qkv",
|
| 94 |
+
"language_model.model.layers.20.mlp.up_proj",
|
| 95 |
+
"language_model.model.layers.20.mlp.down_proj",
|
| 96 |
+
"language_model.model.layers.20.mlp.gate_proj",
|
| 97 |
+
"language_model.model.layers.20.linear_attn.out_proj",
|
| 98 |
+
"language_model.model.layers.20.linear_attn.in_proj_a",
|
| 99 |
+
"language_model.model.layers.20.linear_attn.in_proj_b",
|
| 100 |
+
"language_model.model.layers.20.linear_attn.in_proj_z",
|
| 101 |
+
"language_model.model.layers.20.linear_attn.in_proj_qkv",
|
| 102 |
+
"language_model.model.layers.19.mlp.up_proj",
|
| 103 |
+
"language_model.model.layers.19.mlp.down_proj",
|
| 104 |
+
"language_model.model.layers.19.mlp.gate_proj",
|
| 105 |
+
"language_model.model.layers.19.self_attn.o_proj",
|
| 106 |
+
"language_model.model.layers.19.self_attn.v_proj",
|
| 107 |
+
"language_model.model.layers.19.self_attn.k_proj",
|
| 108 |
+
"language_model.model.layers.19.self_attn.q_proj",
|
| 109 |
+
"language_model.model.layers.18.mlp.up_proj",
|
| 110 |
+
"language_model.model.layers.18.mlp.down_proj",
|
| 111 |
+
"language_model.model.layers.18.mlp.gate_proj",
|
| 112 |
+
"language_model.model.layers.18.linear_attn.out_proj",
|
| 113 |
+
"language_model.model.layers.18.linear_attn.in_proj_a",
|
| 114 |
+
"language_model.model.layers.18.linear_attn.in_proj_b",
|
| 115 |
+
"language_model.model.layers.18.linear_attn.in_proj_z",
|
| 116 |
+
"language_model.model.layers.18.linear_attn.in_proj_qkv",
|
| 117 |
+
"language_model.model.layers.17.mlp.up_proj",
|
| 118 |
+
"language_model.model.layers.17.mlp.down_proj",
|
| 119 |
+
"language_model.model.layers.17.mlp.gate_proj",
|
| 120 |
+
"language_model.model.layers.17.linear_attn.out_proj",
|
| 121 |
+
"language_model.model.layers.17.linear_attn.in_proj_a",
|
| 122 |
+
"language_model.model.layers.17.linear_attn.in_proj_b",
|
| 123 |
+
"language_model.model.layers.17.linear_attn.in_proj_z",
|
| 124 |
+
"language_model.model.layers.17.linear_attn.in_proj_qkv",
|
| 125 |
+
"language_model.model.layers.16.mlp.up_proj",
|
| 126 |
+
"language_model.model.layers.16.mlp.down_proj",
|
| 127 |
+
"language_model.model.layers.16.mlp.gate_proj",
|
| 128 |
+
"language_model.model.layers.16.linear_attn.out_proj",
|
| 129 |
+
"language_model.model.layers.16.linear_attn.in_proj_a",
|
| 130 |
+
"language_model.model.layers.16.linear_attn.in_proj_b",
|
| 131 |
+
"language_model.model.layers.16.linear_attn.in_proj_z",
|
| 132 |
+
"language_model.model.layers.16.linear_attn.in_proj_qkv",
|
| 133 |
+
"language_model.model.layers.15.mlp.up_proj",
|
| 134 |
+
"language_model.model.layers.15.mlp.down_proj",
|
| 135 |
+
"language_model.model.layers.15.mlp.gate_proj",
|
| 136 |
+
"language_model.model.layers.15.self_attn.o_proj",
|
| 137 |
+
"language_model.model.layers.15.self_attn.v_proj",
|
| 138 |
+
"language_model.model.layers.15.self_attn.k_proj",
|
| 139 |
+
"language_model.model.layers.15.self_attn.q_proj",
|
| 140 |
+
"language_model.model.layers.14.mlp.up_proj",
|
| 141 |
+
"language_model.model.layers.14.mlp.down_proj",
|
| 142 |
+
"language_model.model.layers.14.mlp.gate_proj",
|
| 143 |
+
"language_model.model.layers.14.linear_attn.out_proj",
|
| 144 |
+
"language_model.model.layers.14.linear_attn.in_proj_a",
|
| 145 |
+
"language_model.model.layers.14.linear_attn.in_proj_b",
|
| 146 |
+
"language_model.model.layers.14.linear_attn.in_proj_z",
|
| 147 |
+
"language_model.model.layers.14.linear_attn.in_proj_qkv",
|
| 148 |
+
"language_model.model.layers.13.mlp.up_proj",
|
| 149 |
+
"language_model.model.layers.13.mlp.down_proj",
|
| 150 |
+
"language_model.model.layers.13.mlp.gate_proj",
|
| 151 |
+
"language_model.model.layers.13.linear_attn.out_proj",
|
| 152 |
+
"language_model.model.layers.13.linear_attn.in_proj_a",
|
| 153 |
+
"language_model.model.layers.13.linear_attn.in_proj_b",
|
| 154 |
+
"language_model.model.layers.13.linear_attn.in_proj_z",
|
| 155 |
+
"language_model.model.layers.13.linear_attn.in_proj_qkv",
|
| 156 |
+
"language_model.model.layers.12.mlp.up_proj",
|
| 157 |
+
"language_model.model.layers.12.mlp.down_proj",
|
| 158 |
+
"language_model.model.layers.12.mlp.gate_proj",
|
| 159 |
+
"language_model.model.layers.12.linear_attn.out_proj",
|
| 160 |
+
"language_model.model.layers.12.linear_attn.in_proj_a",
|
| 161 |
+
"language_model.model.layers.12.linear_attn.in_proj_b",
|
| 162 |
+
"language_model.model.layers.12.linear_attn.in_proj_z",
|
| 163 |
+
"language_model.model.layers.12.linear_attn.in_proj_qkv",
|
| 164 |
+
"language_model.model.layers.11.mlp.up_proj",
|
| 165 |
+
"language_model.model.layers.11.mlp.down_proj",
|
| 166 |
+
"language_model.model.layers.11.mlp.gate_proj",
|
| 167 |
+
"language_model.model.layers.11.self_attn.o_proj",
|
| 168 |
+
"language_model.model.layers.11.self_attn.v_proj",
|
| 169 |
+
"language_model.model.layers.11.self_attn.k_proj",
|
| 170 |
+
"language_model.model.layers.11.self_attn.q_proj",
|
| 171 |
+
"language_model.model.layers.10.mlp.up_proj",
|
| 172 |
+
"language_model.model.layers.10.mlp.down_proj",
|
| 173 |
+
"language_model.model.layers.10.mlp.gate_proj",
|
| 174 |
+
"language_model.model.layers.10.linear_attn.out_proj",
|
| 175 |
+
"language_model.model.layers.10.linear_attn.in_proj_a",
|
| 176 |
+
"language_model.model.layers.10.linear_attn.in_proj_b",
|
| 177 |
+
"language_model.model.layers.10.linear_attn.in_proj_z",
|
| 178 |
+
"language_model.model.layers.10.linear_attn.in_proj_qkv",
|
| 179 |
+
"language_model.model.layers.9.mlp.up_proj",
|
| 180 |
+
"language_model.model.layers.9.mlp.down_proj",
|
| 181 |
+
"language_model.model.layers.9.mlp.gate_proj",
|
| 182 |
+
"language_model.model.layers.9.linear_attn.out_proj",
|
| 183 |
+
"language_model.model.layers.9.linear_attn.in_proj_a",
|
| 184 |
+
"language_model.model.layers.9.linear_attn.in_proj_b",
|
| 185 |
+
"language_model.model.layers.9.linear_attn.in_proj_z",
|
| 186 |
+
"language_model.model.layers.9.linear_attn.in_proj_qkv",
|
| 187 |
+
"language_model.model.layers.8.mlp.up_proj",
|
| 188 |
+
"language_model.model.layers.8.mlp.down_proj",
|
| 189 |
+
"language_model.model.layers.8.mlp.gate_proj",
|
| 190 |
+
"language_model.model.layers.8.linear_attn.out_proj",
|
| 191 |
+
"language_model.model.layers.8.linear_attn.in_proj_a",
|
| 192 |
+
"language_model.model.layers.8.linear_attn.in_proj_b",
|
| 193 |
+
"language_model.model.layers.8.linear_attn.in_proj_z",
|
| 194 |
+
"language_model.model.layers.8.linear_attn.in_proj_qkv",
|
| 195 |
+
"language_model.model.layers.7.mlp.up_proj",
|
| 196 |
+
"language_model.model.layers.7.mlp.down_proj",
|
| 197 |
+
"language_model.model.layers.7.mlp.gate_proj",
|
| 198 |
+
"language_model.model.layers.7.self_attn.o_proj",
|
| 199 |
+
"language_model.model.layers.7.self_attn.v_proj",
|
| 200 |
+
"language_model.model.layers.7.self_attn.k_proj",
|
| 201 |
+
"language_model.model.layers.7.self_attn.q_proj",
|
| 202 |
+
"language_model.model.layers.6.mlp.up_proj",
|
| 203 |
+
"language_model.model.layers.6.mlp.down_proj",
|
| 204 |
+
"language_model.model.layers.6.mlp.gate_proj",
|
| 205 |
+
"language_model.model.layers.6.linear_attn.out_proj",
|
| 206 |
+
"language_model.model.layers.6.linear_attn.in_proj_a",
|
| 207 |
+
"language_model.model.layers.6.linear_attn.in_proj_b",
|
| 208 |
+
"language_model.model.layers.6.linear_attn.in_proj_z",
|
| 209 |
+
"language_model.model.layers.6.linear_attn.in_proj_qkv",
|
| 210 |
+
"language_model.model.layers.5.mlp.up_proj",
|
| 211 |
+
"language_model.model.layers.5.mlp.down_proj",
|
| 212 |
+
"language_model.model.layers.5.mlp.gate_proj",
|
| 213 |
+
"language_model.model.layers.5.linear_attn.out_proj",
|
| 214 |
+
"language_model.model.layers.5.linear_attn.in_proj_a",
|
| 215 |
+
"language_model.model.layers.5.linear_attn.in_proj_b",
|
| 216 |
+
"language_model.model.layers.5.linear_attn.in_proj_z",
|
| 217 |
+
"language_model.model.layers.5.linear_attn.in_proj_qkv",
|
| 218 |
+
"language_model.model.layers.4.mlp.up_proj",
|
| 219 |
+
"language_model.model.layers.4.mlp.down_proj",
|
| 220 |
+
"language_model.model.layers.4.mlp.gate_proj",
|
| 221 |
+
"language_model.model.layers.4.linear_attn.out_proj",
|
| 222 |
+
"language_model.model.layers.4.linear_attn.in_proj_a",
|
| 223 |
+
"language_model.model.layers.4.linear_attn.in_proj_b",
|
| 224 |
+
"language_model.model.layers.4.linear_attn.in_proj_z",
|
| 225 |
+
"language_model.model.layers.4.linear_attn.in_proj_qkv",
|
| 226 |
+
"language_model.model.layers.3.mlp.up_proj",
|
| 227 |
+
"language_model.model.layers.3.mlp.down_proj",
|
| 228 |
+
"language_model.model.layers.3.mlp.gate_proj",
|
| 229 |
+
"language_model.model.layers.3.self_attn.o_proj",
|
| 230 |
+
"language_model.model.layers.3.self_attn.v_proj",
|
| 231 |
+
"language_model.model.layers.3.self_attn.k_proj",
|
| 232 |
+
"language_model.model.layers.3.self_attn.q_proj",
|
| 233 |
+
"language_model.model.layers.2.mlp.up_proj",
|
| 234 |
+
"language_model.model.layers.2.mlp.down_proj",
|
| 235 |
+
"language_model.model.layers.2.mlp.gate_proj",
|
| 236 |
+
"language_model.model.layers.2.linear_attn.out_proj",
|
| 237 |
+
"language_model.model.layers.2.linear_attn.in_proj_a",
|
| 238 |
+
"language_model.model.layers.2.linear_attn.in_proj_b",
|
| 239 |
+
"language_model.model.layers.2.linear_attn.in_proj_z",
|
| 240 |
+
"language_model.model.layers.2.linear_attn.in_proj_qkv",
|
| 241 |
+
"language_model.model.layers.1.mlp.up_proj",
|
| 242 |
+
"language_model.model.layers.1.mlp.down_proj",
|
| 243 |
+
"language_model.model.layers.1.mlp.gate_proj",
|
| 244 |
+
"language_model.model.layers.1.linear_attn.out_proj",
|
| 245 |
+
"language_model.model.layers.1.linear_attn.in_proj_a",
|
| 246 |
+
"language_model.model.layers.1.linear_attn.in_proj_b",
|
| 247 |
+
"language_model.model.layers.1.linear_attn.in_proj_z",
|
| 248 |
+
"language_model.model.layers.1.linear_attn.in_proj_qkv",
|
| 249 |
+
"language_model.model.layers.0.mlp.up_proj",
|
| 250 |
+
"language_model.model.layers.0.mlp.down_proj",
|
| 251 |
+
"language_model.model.layers.0.mlp.gate_proj",
|
| 252 |
+
"language_model.model.layers.0.linear_attn.out_proj",
|
| 253 |
+
"language_model.model.layers.0.linear_attn.in_proj_a",
|
| 254 |
+
"language_model.model.layers.0.linear_attn.in_proj_b",
|
| 255 |
+
"language_model.model.layers.0.linear_attn.in_proj_z",
|
| 256 |
+
"language_model.model.layers.0.linear_attn.in_proj_qkv"
|
| 257 |
+
]
|
| 258 |
+
}
|
| 259 |
+
}
|
adapter/adapters.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:872135863fbb2c45785d965898d85819efe15052004c76eddf7a2923e9f6f365
|
| 3 |
+
size 64991946
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- set image_count = namespace(value=0) %}
|
| 2 |
+
{%- set video_count = namespace(value=0) %}
|
| 3 |
+
{%- macro render_content(content, do_vision_count, is_system_content=false) %}
|
| 4 |
+
{%- if content is string %}
|
| 5 |
+
{{- content }}
|
| 6 |
+
{%- elif content is iterable and content is not mapping %}
|
| 7 |
+
{%- for item in content %}
|
| 8 |
+
{%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
|
| 9 |
+
{%- if is_system_content %}
|
| 10 |
+
{{- raise_exception('System message cannot contain images.') }}
|
| 11 |
+
{%- endif %}
|
| 12 |
+
{%- if do_vision_count %}
|
| 13 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 14 |
+
{%- endif %}
|
| 15 |
+
{%- if add_vision_id %}
|
| 16 |
+
{{- 'Picture ' ~ image_count.value ~ ': ' }}
|
| 17 |
+
{%- endif %}
|
| 18 |
+
{{- '<|vision_start|><|image_pad|><|vision_end|>' }}
|
| 19 |
+
{%- elif 'video' in item or item.type == 'video' %}
|
| 20 |
+
{%- if is_system_content %}
|
| 21 |
+
{{- raise_exception('System message cannot contain videos.') }}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if do_vision_count %}
|
| 24 |
+
{%- set video_count.value = video_count.value + 1 %}
|
| 25 |
+
{%- endif %}
|
| 26 |
+
{%- if add_vision_id %}
|
| 27 |
+
{{- 'Video ' ~ video_count.value ~ ': ' }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{{- '<|vision_start|><|video_pad|><|vision_end|>' }}
|
| 30 |
+
{%- elif 'text' in item %}
|
| 31 |
+
{{- item.text }}
|
| 32 |
+
{%- else %}
|
| 33 |
+
{{- raise_exception('Unexpected item type in content.') }}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{%- endfor %}
|
| 36 |
+
{%- elif content is none or content is undefined %}
|
| 37 |
+
{{- '' }}
|
| 38 |
+
{%- else %}
|
| 39 |
+
{{- raise_exception('Unexpected content type.') }}
|
| 40 |
+
{%- endif %}
|
| 41 |
+
{%- endmacro %}
|
| 42 |
+
{%- if not messages %}
|
| 43 |
+
{{- raise_exception('No messages provided.') }}
|
| 44 |
+
{%- endif %}
|
| 45 |
+
{%- if tools and tools is iterable and tools is not mapping %}
|
| 46 |
+
{{- '<|im_start|>system\n' }}
|
| 47 |
+
{{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
|
| 48 |
+
{%- for tool in tools %}
|
| 49 |
+
{{- "\n" }}
|
| 50 |
+
{{- tool | tojson }}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{{- "\n</tools>" }}
|
| 53 |
+
{{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
|
| 54 |
+
{%- if messages[0].role == 'system' %}
|
| 55 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 56 |
+
{%- if content %}
|
| 57 |
+
{{- '\n\n' + content }}
|
| 58 |
+
{%- endif %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<|im_end|>\n' }}
|
| 61 |
+
{%- else %}
|
| 62 |
+
{%- if messages[0].role == 'system' %}
|
| 63 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 64 |
+
{{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
|
| 65 |
+
{%- endif %}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 68 |
+
{%- for message in messages[::-1] %}
|
| 69 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 70 |
+
{%- if ns.multi_step_tool and message.role == "user" %}
|
| 71 |
+
{%- set content = render_content(message.content, false)|trim %}
|
| 72 |
+
{%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
|
| 73 |
+
{%- set ns.multi_step_tool = false %}
|
| 74 |
+
{%- set ns.last_query_index = index %}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{%- endif %}
|
| 77 |
+
{%- endfor %}
|
| 78 |
+
{%- if ns.multi_step_tool %}
|
| 79 |
+
{{- raise_exception('No user query found in messages.') }}
|
| 80 |
+
{%- endif %}
|
| 81 |
+
{%- for message in messages %}
|
| 82 |
+
{%- set content = render_content(message.content, true)|trim %}
|
| 83 |
+
{%- if message.role == "system" %}
|
| 84 |
+
{%- if not loop.first %}
|
| 85 |
+
{{- raise_exception('System message must be at the beginning.') }}
|
| 86 |
+
{%- endif %}
|
| 87 |
+
{%- elif message.role == "user" %}
|
| 88 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 89 |
+
{%- elif message.role == "assistant" %}
|
| 90 |
+
{%- set reasoning_content = '' %}
|
| 91 |
+
{%- if message.reasoning_content is string %}
|
| 92 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 93 |
+
{%- else %}
|
| 94 |
+
{%- if '</think>' in content %}
|
| 95 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 96 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 97 |
+
{%- endif %}
|
| 98 |
+
{%- endif %}
|
| 99 |
+
{%- set reasoning_content = reasoning_content|trim %}
|
| 100 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 101 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
|
| 102 |
+
{%- else %}
|
| 103 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 104 |
+
{%- endif %}
|
| 105 |
+
{%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
|
| 106 |
+
{%- for tool_call in message.tool_calls %}
|
| 107 |
+
{%- if tool_call.function is defined %}
|
| 108 |
+
{%- set tool_call = tool_call.function %}
|
| 109 |
+
{%- endif %}
|
| 110 |
+
{%- if loop.first %}
|
| 111 |
+
{%- if content|trim %}
|
| 112 |
+
{{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 113 |
+
{%- else %}
|
| 114 |
+
{{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 115 |
+
{%- endif %}
|
| 116 |
+
{%- else %}
|
| 117 |
+
{{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
|
| 118 |
+
{%- endif %}
|
| 119 |
+
{%- if tool_call.arguments is defined %}
|
| 120 |
+
{%- for args_name, args_value in tool_call.arguments|items %}
|
| 121 |
+
{{- '<parameter=' + args_name + '>\n' }}
|
| 122 |
+
{%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
|
| 123 |
+
{{- args_value }}
|
| 124 |
+
{{- '\n</parameter>\n' }}
|
| 125 |
+
{%- endfor %}
|
| 126 |
+
{%- endif %}
|
| 127 |
+
{{- '</function>\n</tool_call>' }}
|
| 128 |
+
{%- endfor %}
|
| 129 |
+
{%- endif %}
|
| 130 |
+
{{- '<|im_end|>\n' }}
|
| 131 |
+
{%- elif message.role == "tool" %}
|
| 132 |
+
{%- if loop.previtem and loop.previtem.role != "tool" %}
|
| 133 |
+
{{- '<|im_start|>user' }}
|
| 134 |
+
{%- endif %}
|
| 135 |
+
{{- '\n<tool_response>\n' }}
|
| 136 |
+
{{- content }}
|
| 137 |
+
{{- '\n</tool_response>' }}
|
| 138 |
+
{%- if not loop.last and loop.nextitem.role != "tool" %}
|
| 139 |
+
{{- '<|im_end|>\n' }}
|
| 140 |
+
{%- elif loop.last %}
|
| 141 |
+
{{- '<|im_end|>\n' }}
|
| 142 |
+
{%- endif %}
|
| 143 |
+
{%- else %}
|
| 144 |
+
{{- raise_exception('Unexpected message role.') }}
|
| 145 |
+
{%- endif %}
|
| 146 |
+
{%- endfor %}
|
| 147 |
+
{%- if add_generation_prompt %}
|
| 148 |
+
{{- '<|im_start|>assistant\n' }}
|
| 149 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 150 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 151 |
+
{%- else %}
|
| 152 |
+
{{- '<think>\n' }}
|
| 153 |
+
{%- endif %}
|
| 154 |
+
{%- endif %}
|
config.json
ADDED
|
@@ -0,0 +1,104 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"image_token_id": 248056,
|
| 6 |
+
"model_type": "qwen3_5",
|
| 7 |
+
"text_config": {
|
| 8 |
+
"attention_bias": false,
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"attn_output_gate": true,
|
| 11 |
+
"dtype": "bfloat16",
|
| 12 |
+
"eos_token_id": 248044,
|
| 13 |
+
"full_attention_interval": 4,
|
| 14 |
+
"head_dim": 256,
|
| 15 |
+
"hidden_act": "silu",
|
| 16 |
+
"hidden_size": 2560,
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 9216,
|
| 19 |
+
"layer_types": [
|
| 20 |
+
"linear_attention",
|
| 21 |
+
"linear_attention",
|
| 22 |
+
"linear_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"linear_attention",
|
| 25 |
+
"linear_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"linear_attention",
|
| 29 |
+
"linear_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"linear_attention",
|
| 33 |
+
"linear_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"linear_attention",
|
| 37 |
+
"linear_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"linear_attention",
|
| 41 |
+
"linear_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"linear_attention",
|
| 45 |
+
"linear_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"linear_attention",
|
| 49 |
+
"linear_attention",
|
| 50 |
+
"linear_attention",
|
| 51 |
+
"full_attention"
|
| 52 |
+
],
|
| 53 |
+
"linear_conv_kernel_dim": 4,
|
| 54 |
+
"linear_key_head_dim": 128,
|
| 55 |
+
"linear_num_key_heads": 16,
|
| 56 |
+
"linear_num_value_heads": 32,
|
| 57 |
+
"linear_value_head_dim": 128,
|
| 58 |
+
"max_position_embeddings": 262144,
|
| 59 |
+
"mlp_only_layers": [],
|
| 60 |
+
"model_type": "qwen3_5_text",
|
| 61 |
+
"mtp_num_hidden_layers": 1,
|
| 62 |
+
"mtp_use_dedicated_embeddings": false,
|
| 63 |
+
"num_attention_heads": 16,
|
| 64 |
+
"num_hidden_layers": 32,
|
| 65 |
+
"num_key_value_heads": 4,
|
| 66 |
+
"rms_norm_eps": 1e-06,
|
| 67 |
+
"tie_word_embeddings": true,
|
| 68 |
+
"use_cache": true,
|
| 69 |
+
"vocab_size": 248320,
|
| 70 |
+
"mamba_ssm_dtype": "float32",
|
| 71 |
+
"rope_parameters": {
|
| 72 |
+
"mrope_interleaved": true,
|
| 73 |
+
"mrope_section": [
|
| 74 |
+
11,
|
| 75 |
+
11,
|
| 76 |
+
10
|
| 77 |
+
],
|
| 78 |
+
"rope_type": "default",
|
| 79 |
+
"rope_theta": 10000000,
|
| 80 |
+
"partial_rotary_factor": 0.25
|
| 81 |
+
}
|
| 82 |
+
},
|
| 83 |
+
"tie_word_embeddings": true,
|
| 84 |
+
"transformers_version": "4.57.0.dev0",
|
| 85 |
+
"video_token_id": 248057,
|
| 86 |
+
"vision_config": {
|
| 87 |
+
"deepstack_visual_indexes": [],
|
| 88 |
+
"depth": 24,
|
| 89 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 90 |
+
"hidden_size": 1024,
|
| 91 |
+
"in_channels": 3,
|
| 92 |
+
"initializer_range": 0.02,
|
| 93 |
+
"intermediate_size": 4096,
|
| 94 |
+
"model_type": "qwen3_5",
|
| 95 |
+
"num_heads": 16,
|
| 96 |
+
"num_position_embeddings": 2304,
|
| 97 |
+
"out_hidden_size": 2560,
|
| 98 |
+
"patch_size": 16,
|
| 99 |
+
"spatial_merge_size": 2,
|
| 100 |
+
"temporal_patch_size": 2
|
| 101 |
+
},
|
| 102 |
+
"vision_end_token_id": 248054,
|
| 103 |
+
"vision_start_token_id": 248053
|
| 104 |
+
}
|
inference/cua_decide.py
ADDED
|
@@ -0,0 +1,318 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"""Typed next-step decisions on a macOS screenshot with the macOS CUA decision model.
|
| 2 |
+
|
| 3 |
+
The model does not generate free text. Each step is asked as up to three multiple-choice
|
| 4 |
+
questions, and every option's probability is read from one shared forward pass:
|
| 5 |
+
|
| 6 |
+
- action: which kind of action comes next (click, double-click, ..., finish);
|
| 7 |
+
- next_target: which on-screen element the next action acts on;
|
| 8 |
+
- grounding: which element matches a description you give.
|
| 9 |
+
|
| 10 |
+
Elements are candidates found on the screenshot: text lines from the macOS Vision
|
| 11 |
+
framework, plus, optionally, icons from OmniParser v2's detector (AGPL-3.0 weights, not
|
| 12 |
+
bundled; see the model card). The prompt format below is the one the model was trained on;
|
| 13 |
+
changing it changes the answers.
|
| 14 |
+
|
| 15 |
+
pip install "mlx-vlm>=0.7,<0.8" pyobjc-framework-vision pyobjc-framework-quartz
|
| 16 |
+
python cua_decide.py --model <repo or dir> --image screen.png \\
|
| 17 |
+
--task "Turn on Dark Mode" --past "Click the Apple menu" \\
|
| 18 |
+
--temperatures temperatures.json
|
| 19 |
+
|
| 20 |
+
This file is self-contained and mirrors the code the model was trained and evaluated with.
|
| 21 |
+
"""
|
| 22 |
+
|
| 23 |
+
from __future__ import annotations
|
| 24 |
+
|
| 25 |
+
import argparse
|
| 26 |
+
import json
|
| 27 |
+
import math
|
| 28 |
+
from dataclasses import dataclass
|
| 29 |
+
from pathlib import Path
|
| 30 |
+
from typing import Any
|
| 31 |
+
|
| 32 |
+
SYSTEM = ("You are a macOS computer-use assistant. You look at the current screen and "
|
| 33 |
+
"answer each question by choosing one of the options it offers.")
|
| 34 |
+
# Option label shown to the model, and the action it stands for.
|
| 35 |
+
ACTIONS = [("click", "click"), ("double-click", "doubleClick"),
|
| 36 |
+
("right-click", "rightClick"), ("drag", "dragTo"), ("type", "write"),
|
| 37 |
+
("press", "press"), ("shortcut", "hotkey"), ("scroll", "scroll"),
|
| 38 |
+
("hover", "moveTo"), ("wait", "wait"), ("finish", "finish")]
|
| 39 |
+
|
| 40 |
+
|
| 41 |
+
# -- candidates ------------------------------------------------------------------------------
|
| 42 |
+
|
| 43 |
+
@dataclass(frozen=True)
|
| 44 |
+
class Candidate:
|
| 45 |
+
"""A box on the screenshot, normalized to [0, 1] from the top left."""
|
| 46 |
+
x1: float
|
| 47 |
+
y1: float
|
| 48 |
+
x2: float
|
| 49 |
+
y2: float
|
| 50 |
+
text: str
|
| 51 |
+
source: str
|
| 52 |
+
|
| 53 |
+
@property
|
| 54 |
+
def centre(self) -> tuple[float, float]:
|
| 55 |
+
return (self.x1 + self.x2) / 2, (self.y1 + self.y2) / 2
|
| 56 |
+
|
| 57 |
+
|
| 58 |
+
def text_candidates(image: Path, min_height: float = 0.006) -> list[Candidate]:
|
| 59 |
+
"""Every text line the macOS Vision framework recognises."""
|
| 60 |
+
import Vision
|
| 61 |
+
from Foundation import NSURL
|
| 62 |
+
|
| 63 |
+
handler = Vision.VNImageRequestHandler.alloc().initWithURL_options_(
|
| 64 |
+
NSURL.fileURLWithPath_(str(image)), None)
|
| 65 |
+
request = Vision.VNRecognizeTextRequest.alloc().init()
|
| 66 |
+
request.setRecognitionLevel_(Vision.VNRequestTextRecognitionLevelAccurate)
|
| 67 |
+
request.setUsesLanguageCorrection_(False)
|
| 68 |
+
ok, error = handler.performRequests_error_([request], None)
|
| 69 |
+
if not ok:
|
| 70 |
+
raise RuntimeError(f"text recognition failed on {image}: {error}")
|
| 71 |
+
found = []
|
| 72 |
+
for observation in request.results() or []:
|
| 73 |
+
top = observation.topCandidates_(1)
|
| 74 |
+
if not top:
|
| 75 |
+
continue
|
| 76 |
+
box = observation.boundingBox()
|
| 77 |
+
x1, width = float(box.origin.x), float(box.size.width)
|
| 78 |
+
y2 = 1.0 - float(box.origin.y)
|
| 79 |
+
y1 = y2 - float(box.size.height)
|
| 80 |
+
if y2 - y1 >= min_height:
|
| 81 |
+
found.append(Candidate(max(0.0, x1), max(0.0, y1), min(1.0, x1 + width),
|
| 82 |
+
min(1.0, y2), str(top[0].string()), "text"))
|
| 83 |
+
return found
|
| 84 |
+
|
| 85 |
+
|
| 86 |
+
def icon_candidates(image: Path, weights: Path, confidence: float = 0.05) -> list[Candidate]:
|
| 87 |
+
"""Interactable regions from OmniParser v2's icon detector (you supply its weights)."""
|
| 88 |
+
from ultralytics import YOLO
|
| 89 |
+
|
| 90 |
+
result = YOLO(str(weights)).predict(str(image), conf=confidence, iou=0.3, imgsz=1280,
|
| 91 |
+
verbose=False)[0]
|
| 92 |
+
width, height = result.orig_shape[1], result.orig_shape[0]
|
| 93 |
+
return [Candidate(x1 / width, y1 / height, x2 / width, y2 / height, "", "icon")
|
| 94 |
+
for x1, y1, x2, y2 in result.boxes.xyxy.tolist()]
|
| 95 |
+
|
| 96 |
+
|
| 97 |
+
def _overlap(first: Candidate, second: Candidate) -> float:
|
| 98 |
+
width = min(first.x2, second.x2) - max(first.x1, second.x1)
|
| 99 |
+
height = min(first.y2, second.y2) - max(first.y1, second.y1)
|
| 100 |
+
if width <= 0 or height <= 0:
|
| 101 |
+
return 0.0
|
| 102 |
+
smaller = min((first.x2 - first.x1) * (first.y2 - first.y1),
|
| 103 |
+
(second.x2 - second.x1) * (second.y2 - second.y1))
|
| 104 |
+
return width * height / smaller if smaller else 0.0
|
| 105 |
+
|
| 106 |
+
|
| 107 |
+
def screen_candidates(image: Path, icon_weights: Path | None = None) -> list[Candidate]:
|
| 108 |
+
"""Text lines, then detected icons no text line accounts for, in reading order."""
|
| 109 |
+
texts = text_candidates(image)
|
| 110 |
+
icons = [] if icon_weights is None else [
|
| 111 |
+
icon for icon in icon_candidates(image, icon_weights)
|
| 112 |
+
if all(_overlap(icon, text) < 0.6 for text in texts)]
|
| 113 |
+
return sorted(texts + icons, key=lambda item: (round(item.y1, 2), item.x1))
|
| 114 |
+
|
| 115 |
+
|
| 116 |
+
# -- prompt ---------------------------------------------------------------------------------
|
| 117 |
+
|
| 118 |
+
def label_width(found: list[Candidate]) -> int:
|
| 119 |
+
return max(2, len(str(max(0, len(found) - 1))))
|
| 120 |
+
|
| 121 |
+
|
| 122 |
+
def context(task: str, past_actions: list[str], found: list[Candidate]) -> str:
|
| 123 |
+
"""The shared part of every question: task, the last three actions, the elements."""
|
| 124 |
+
width = label_width(found)
|
| 125 |
+
elements = "\n".join(
|
| 126 |
+
f"[{index:0{width}d}] at ({round(c.centre[0] * 1000)}, {round(c.centre[1] * 1000)}) "
|
| 127 |
+
+ (f'"{c.text[:60]}"' if c.text else "icon or control")
|
| 128 |
+
for index, c in enumerate(found))
|
| 129 |
+
history = "\n".join(past_actions[-3:]) or "(none)"
|
| 130 |
+
return (f"Task: {task}\nPast actions:\n{history}\n\n"
|
| 131 |
+
f"Elements on the current screen, numbered, with centre coordinates "
|
| 132 |
+
f"from 0 to 1000:\n{elements}")
|
| 133 |
+
|
| 134 |
+
|
| 135 |
+
def questions(found: list[Candidate], describe: str | None) -> list[tuple[str, str, list[str]]]:
|
| 136 |
+
"""(name, question text, option labels) for each question to ask."""
|
| 137 |
+
width = label_width(found)
|
| 138 |
+
labels = [f"{index:0{width}d}" for index in range(len(found))]
|
| 139 |
+
asked = [("action", "What kind of action comes next? Answer with exactly one word from: "
|
| 140 |
+
+ ", ".join(label for label, _ in ACTIONS) + ".", [label for label, _ in ACTIONS])]
|
| 141 |
+
if found:
|
| 142 |
+
asked.append(("next_target", "Which element does the next action act on? "
|
| 143 |
+
f"Answer with the element's {width}-digit number.", labels))
|
| 144 |
+
if describe:
|
| 145 |
+
asked.append(("grounding", f"Which element is this: {describe}? "
|
| 146 |
+
f"Answer with the element's {width}-digit number.", labels))
|
| 147 |
+
return asked
|
| 148 |
+
|
| 149 |
+
|
| 150 |
+
# -- scoring --------------------------------------------------------------------------------
|
| 151 |
+
|
| 152 |
+
class DecisionModel:
|
| 153 |
+
"""Reads option probabilities from one shared forward pass over the common prefix."""
|
| 154 |
+
|
| 155 |
+
def __init__(self, path: str, adapter: str | None = None) -> None:
|
| 156 |
+
import mlx.core as mx
|
| 157 |
+
import mlx_vlm
|
| 158 |
+
from mlx_vlm import utils
|
| 159 |
+
from mlx_vlm.models import cache
|
| 160 |
+
|
| 161 |
+
self.mx, self.utils, self.cache = mx, utils, cache
|
| 162 |
+
self.model, self.processor = mlx_vlm.load(path, adapter_path=adapter, lazy=False)
|
| 163 |
+
# The resolution the model was trained at.
|
| 164 |
+
self.processor.image_processor.max_pixels = 786432
|
| 165 |
+
self.processor.image_processor.min_pixels = 65536
|
| 166 |
+
self.model.eval()
|
| 167 |
+
self.tokenizer = self.processor.tokenizer
|
| 168 |
+
mx.set_cache_limit(2 * 1024**3)
|
| 169 |
+
|
| 170 |
+
def _prompt(self, text: str) -> str:
|
| 171 |
+
messages = [{"role": "system", "content": SYSTEM},
|
| 172 |
+
{"role": "user", "content": [{"type": "image"},
|
| 173 |
+
{"type": "text", "text": text}]}]
|
| 174 |
+
return self.processor.apply_chat_template(messages, tokenize=False,
|
| 175 |
+
add_generation_prompt=True,
|
| 176 |
+
enable_thinking=False)
|
| 177 |
+
|
| 178 |
+
def _reset(self) -> None:
|
| 179 |
+
# Positions carried over from the last prompt must not leak into this one.
|
| 180 |
+
for name in ("_position_ids", "_rope_deltas"):
|
| 181 |
+
if hasattr(self.model.language_model, name):
|
| 182 |
+
setattr(self.model.language_model, name, None)
|
| 183 |
+
|
| 184 |
+
@staticmethod
|
| 185 |
+
def _snapshot(entries: list[Any]) -> list[Any]:
|
| 186 |
+
return [("kv", entry.offset) if hasattr(entry, "offset") else ("arrays", list(entry.cache))
|
| 187 |
+
for entry in entries]
|
| 188 |
+
|
| 189 |
+
@staticmethod
|
| 190 |
+
def _restore(entries: list[Any], state: list[Any]) -> None:
|
| 191 |
+
for entry, (kind, value) in zip(entries, state, strict=True):
|
| 192 |
+
if kind == "kv":
|
| 193 |
+
entry.offset = value
|
| 194 |
+
else:
|
| 195 |
+
entry.cache = list(value)
|
| 196 |
+
|
| 197 |
+
def _next(self, entries: list[Any], tokens: list[int], kwargs: dict[str, Any]) -> Any:
|
| 198 |
+
mx = self.mx
|
| 199 |
+
out = self.model.language_model(mx.array([tokens]), cache=entries, **kwargs)
|
| 200 |
+
logits = out.logits[:, -1, :].astype(mx.float32)
|
| 201 |
+
return (logits - mx.logsumexp(logits, axis=-1, keepdims=True))[0]
|
| 202 |
+
|
| 203 |
+
def decide(self, image: Path, shared: str,
|
| 204 |
+
asked: list[tuple[str, str, list[str]]]) -> dict[str, list[float]]:
|
| 205 |
+
"""Each question's option probabilities, from one pass over the shared prefix."""
|
| 206 |
+
mx = self.mx
|
| 207 |
+
lm = self.model.language_model
|
| 208 |
+
encoded = [self.utils.prepare_inputs(self.processor, images=[str(image)],
|
| 209 |
+
prompts=[self._prompt(f"{shared}\n\n{text}")],
|
| 210 |
+
image_token_index=self.model.config.image_token_index)
|
| 211 |
+
for _, text, _ in asked]
|
| 212 |
+
sequences = [inputs["input_ids"][0].tolist() for inputs in encoded]
|
| 213 |
+
common = 0
|
| 214 |
+
while all(len(seq) > common + 1 for seq in sequences) and len(
|
| 215 |
+
{seq[common] for seq in sequences}) == 1:
|
| 216 |
+
common += 1
|
| 217 |
+
self._reset()
|
| 218 |
+
entries = self.cache.make_prompt_cache(lm)
|
| 219 |
+
first = encoded[0]
|
| 220 |
+
prefix = first["input_ids"][:, :common]
|
| 221 |
+
extra = {key: value for key, value in first.items()
|
| 222 |
+
if key not in ("input_ids", "pixel_values", "attention_mask")}
|
| 223 |
+
embedded = self.model.get_input_embeddings(
|
| 224 |
+
prefix, first.get("pixel_values"), mask=first["attention_mask"][:, :common], **extra)
|
| 225 |
+
kwargs = {key: value for key, value in embedded.to_dict().items()
|
| 226 |
+
if key != "inputs_embeds" and value is not None}
|
| 227 |
+
lm(prefix, inputs_embeds=embedded.inputs_embeds, cache=entries, skip_logits=True,
|
| 228 |
+
return_hidden=True, **kwargs)
|
| 229 |
+
# Positions are only right for the prefix; later tokens continue from the cache.
|
| 230 |
+
kwargs = {key: value for key, value in kwargs.items() if key != "position_ids"}
|
| 231 |
+
mx.eval([entry.state for entry in entries])
|
| 232 |
+
base = self._snapshot(entries)
|
| 233 |
+
result = {}
|
| 234 |
+
for (name, _, labels), sequence in zip(asked, sequences, strict=True):
|
| 235 |
+
root = self._next(entries, sequence[common:], kwargs)
|
| 236 |
+
after_question = self._snapshot(entries)
|
| 237 |
+
tokens = [self.tokenizer.encode(label, add_special_tokens=False) for label in labels]
|
| 238 |
+
scores = self._walk(entries, root, tokens, kwargs, after_question)
|
| 239 |
+
top = max(scores)
|
| 240 |
+
weights = [math.exp(score - top) for score in scores]
|
| 241 |
+
total = sum(weights)
|
| 242 |
+
result[name] = [weight / total for weight in weights]
|
| 243 |
+
self._restore(entries, base)
|
| 244 |
+
mx.clear_cache()
|
| 245 |
+
return result
|
| 246 |
+
|
| 247 |
+
def _walk(self, entries: list[Any], root: Any, labels: list[list[int]],
|
| 248 |
+
kwargs: dict[str, Any], root_state: list[Any]) -> list[float]:
|
| 249 |
+
"""Log probability of every label, one step per distinct token prefix."""
|
| 250 |
+
scores = [0.0] * len(labels)
|
| 251 |
+
pending = [((), root, root_state)]
|
| 252 |
+
while pending:
|
| 253 |
+
prefix, logprobs, state = pending.pop()
|
| 254 |
+
depth = len(prefix)
|
| 255 |
+
members = [index for index, tokens in enumerate(labels)
|
| 256 |
+
if tuple(tokens[:depth]) == prefix and len(tokens) > depth]
|
| 257 |
+
if not members:
|
| 258 |
+
continue
|
| 259 |
+
row = logprobs.tolist()
|
| 260 |
+
children: dict[int, list[int]] = {}
|
| 261 |
+
for index in members:
|
| 262 |
+
token = labels[index][depth]
|
| 263 |
+
scores[index] += row[token]
|
| 264 |
+
children.setdefault(token, []).append(index)
|
| 265 |
+
for token, group in children.items():
|
| 266 |
+
if all(len(labels[index]) == depth + 1 for index in group):
|
| 267 |
+
continue
|
| 268 |
+
self._restore(entries, state)
|
| 269 |
+
following = self._next(entries, [token], kwargs)
|
| 270 |
+
pending.append((prefix + (token,), following, self._snapshot(entries)))
|
| 271 |
+
return scores
|
| 272 |
+
|
| 273 |
+
|
| 274 |
+
def calibrated(probabilities: list[float], temperature: float) -> list[float]:
|
| 275 |
+
"""Temperature scaling: the same ranking, confidence that means what it says."""
|
| 276 |
+
logs = [math.log(max(value, 1e-12)) / temperature for value in probabilities]
|
| 277 |
+
top = max(logs)
|
| 278 |
+
weights = [math.exp(value - top) for value in logs]
|
| 279 |
+
total = sum(weights)
|
| 280 |
+
return [weight / total for weight in weights]
|
| 281 |
+
|
| 282 |
+
|
| 283 |
+
def main() -> None:
|
| 284 |
+
parser = argparse.ArgumentParser(description=__doc__,
|
| 285 |
+
formatter_class=argparse.RawDescriptionHelpFormatter)
|
| 286 |
+
parser.add_argument("--model", required=True, help="merged model repo/dir, or the base")
|
| 287 |
+
parser.add_argument("--adapter", help="LoRA adapter dir, when --model is the base")
|
| 288 |
+
parser.add_argument("--image", type=Path, required=True)
|
| 289 |
+
parser.add_argument("--task", required=True)
|
| 290 |
+
parser.add_argument("--past", action="append", default=[], help="a past action, in order")
|
| 291 |
+
parser.add_argument("--describe", help="also ground this element description")
|
| 292 |
+
parser.add_argument("--icon-weights", type=Path,
|
| 293 |
+
help="OmniParser v2 icon_detect/model.pt (AGPL-3.0, download yourself)")
|
| 294 |
+
parser.add_argument("--temperatures", type=Path, help="temperatures.json from the repo")
|
| 295 |
+
args = parser.parse_args()
|
| 296 |
+
found = screen_candidates(args.image, args.icon_weights)
|
| 297 |
+
asked = questions(found, args.describe)
|
| 298 |
+
probabilities = DecisionModel(args.model, args.adapter).decide(
|
| 299 |
+
args.image, context(args.task, args.past, found), asked)
|
| 300 |
+
temperatures = json.loads(args.temperatures.read_text())["questions"] \
|
| 301 |
+
if args.temperatures else {}
|
| 302 |
+
for name, _, labels in asked:
|
| 303 |
+
scaled = calibrated(probabilities[name], temperatures.get(name, {}).get("temperature", 1.0))
|
| 304 |
+
best = max(range(len(scaled)), key=scaled.__getitem__)
|
| 305 |
+
answer: dict[str, Any] = {"question": name, "choice": labels[best],
|
| 306 |
+
"confidence": round(scaled[best], 4)}
|
| 307 |
+
if name == "action":
|
| 308 |
+
answer["action"] = ACTIONS[best][1]
|
| 309 |
+
else:
|
| 310 |
+
element = found[best]
|
| 311 |
+
answer["element"] = {"text": element.text, "centre": [round(v, 4) for v in element.centre],
|
| 312 |
+
"box": [round(v, 4) for v in (element.x1, element.y1,
|
| 313 |
+
element.x2, element.y2)]}
|
| 314 |
+
print(json.dumps(answer, ensure_ascii=False))
|
| 315 |
+
|
| 316 |
+
|
| 317 |
+
if __name__ == "__main__":
|
| 318 |
+
main()
|
merge_check.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"fused_layers": 248,
|
| 3 |
+
"check_steps": 5,
|
| 4 |
+
"worst_total_variation_vs_adapter": 0.04049
|
| 5 |
+
}
|
model-00001-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:83a8837096b98320146ea9705130cb70e9ee445bc74ae04f1338985b54a75687
|
| 3 |
+
size 5366980383
|
model-00002-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b6428660ef6d18286f9a92ef85263f178189eb4c099541ebc1b964f73818f18
|
| 3 |
+
size 3711641426
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,730 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_size": 9078532608
|
| 4 |
+
},
|
| 5 |
+
"weight_map": {
|
| 6 |
+
"language_model.model.embed_tokens.weight": "model-00001-of-00002.safetensors",
|
| 7 |
+
"language_model.model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 8 |
+
"language_model.model.layers.0.linear_attn.A_log": "model-00001-of-00002.safetensors",
|
| 9 |
+
"language_model.model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
|
| 10 |
+
"language_model.model.layers.0.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
|
| 11 |
+
"language_model.model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
|
| 12 |
+
"language_model.model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
|
| 13 |
+
"language_model.model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
|
| 14 |
+
"language_model.model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
|
| 15 |
+
"language_model.model.layers.0.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
|
| 16 |
+
"language_model.model.layers.0.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
|
| 17 |
+
"language_model.model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 18 |
+
"language_model.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 19 |
+
"language_model.model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 20 |
+
"language_model.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 21 |
+
"language_model.model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 22 |
+
"language_model.model.layers.1.linear_attn.A_log": "model-00001-of-00002.safetensors",
|
| 23 |
+
"language_model.model.layers.1.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
|
| 24 |
+
"language_model.model.layers.1.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
|
| 25 |
+
"language_model.model.layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
|
| 26 |
+
"language_model.model.layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
|
| 27 |
+
"language_model.model.layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
|
| 28 |
+
"language_model.model.layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
|
| 29 |
+
"language_model.model.layers.1.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
|
| 30 |
+
"language_model.model.layers.1.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
|
| 31 |
+
"language_model.model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 32 |
+
"language_model.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 33 |
+
"language_model.model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 34 |
+
"language_model.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 35 |
+
"language_model.model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 36 |
+
"language_model.model.layers.10.linear_attn.A_log": "model-00001-of-00002.safetensors",
|
| 37 |
+
"language_model.model.layers.10.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
|
| 38 |
+
"language_model.model.layers.10.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
|
| 39 |
+
"language_model.model.layers.10.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
|
| 40 |
+
"language_model.model.layers.10.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
|
| 41 |
+
"language_model.model.layers.10.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
|
| 42 |
+
"language_model.model.layers.10.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
|
| 43 |
+
"language_model.model.layers.10.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
|
| 44 |
+
"language_model.model.layers.10.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
|
| 45 |
+
"language_model.model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 46 |
+
"language_model.model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 47 |
+
"language_model.model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 48 |
+
"language_model.model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 49 |
+
"language_model.model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 50 |
+
"language_model.model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 51 |
+
"language_model.model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 52 |
+
"language_model.model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 53 |
+
"language_model.model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 54 |
+
"language_model.model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 55 |
+
"language_model.model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 56 |
+
"language_model.model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 57 |
+
"language_model.model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 58 |
+
"language_model.model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 59 |
+
"language_model.model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 60 |
+
"language_model.model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 61 |
+
"language_model.model.layers.12.linear_attn.A_log": "model-00001-of-00002.safetensors",
|
| 62 |
+
"language_model.model.layers.12.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
|
| 63 |
+
"language_model.model.layers.12.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
|
| 64 |
+
"language_model.model.layers.12.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
|
| 65 |
+
"language_model.model.layers.12.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
|
| 66 |
+
"language_model.model.layers.12.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
|
| 67 |
+
"language_model.model.layers.12.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
|
| 68 |
+
"language_model.model.layers.12.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
|
| 69 |
+
"language_model.model.layers.12.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
|
| 70 |
+
"language_model.model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 71 |
+
"language_model.model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 72 |
+
"language_model.model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 73 |
+
"language_model.model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 74 |
+
"language_model.model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 75 |
+
"language_model.model.layers.13.linear_attn.A_log": "model-00001-of-00002.safetensors",
|
| 76 |
+
"language_model.model.layers.13.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
|
| 77 |
+
"language_model.model.layers.13.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
|
| 78 |
+
"language_model.model.layers.13.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
|
| 79 |
+
"language_model.model.layers.13.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
|
| 80 |
+
"language_model.model.layers.13.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
|
| 81 |
+
"language_model.model.layers.13.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
|
| 82 |
+
"language_model.model.layers.13.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
|
| 83 |
+
"language_model.model.layers.13.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
|
| 84 |
+
"language_model.model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 85 |
+
"language_model.model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 86 |
+
"language_model.model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 87 |
+
"language_model.model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 88 |
+
"language_model.model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 89 |
+
"language_model.model.layers.14.linear_attn.A_log": "model-00001-of-00002.safetensors",
|
| 90 |
+
"language_model.model.layers.14.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
|
| 91 |
+
"language_model.model.layers.14.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
|
| 92 |
+
"language_model.model.layers.14.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
|
| 93 |
+
"language_model.model.layers.14.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
|
| 94 |
+
"language_model.model.layers.14.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
|
| 95 |
+
"language_model.model.layers.14.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
|
| 96 |
+
"language_model.model.layers.14.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
|
| 97 |
+
"language_model.model.layers.14.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
|
| 98 |
+
"language_model.model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 99 |
+
"language_model.model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 100 |
+
"language_model.model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 101 |
+
"language_model.model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 102 |
+
"language_model.model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 103 |
+
"language_model.model.layers.15.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 104 |
+
"language_model.model.layers.15.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 105 |
+
"language_model.model.layers.15.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 106 |
+
"language_model.model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 107 |
+
"language_model.model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 108 |
+
"language_model.model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 109 |
+
"language_model.model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 110 |
+
"language_model.model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 111 |
+
"language_model.model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 112 |
+
"language_model.model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 113 |
+
"language_model.model.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 114 |
+
"language_model.model.layers.16.linear_attn.A_log": "model-00002-of-00002.safetensors",
|
| 115 |
+
"language_model.model.layers.16.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
|
| 116 |
+
"language_model.model.layers.16.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
|
| 117 |
+
"language_model.model.layers.16.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
|
| 118 |
+
"language_model.model.layers.16.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
|
| 119 |
+
"language_model.model.layers.16.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
|
| 120 |
+
"language_model.model.layers.16.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
|
| 121 |
+
"language_model.model.layers.16.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
|
| 122 |
+
"language_model.model.layers.16.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
|
| 123 |
+
"language_model.model.layers.16.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 124 |
+
"language_model.model.layers.16.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 125 |
+
"language_model.model.layers.16.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 126 |
+
"language_model.model.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 127 |
+
"language_model.model.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 128 |
+
"language_model.model.layers.17.linear_attn.A_log": "model-00002-of-00002.safetensors",
|
| 129 |
+
"language_model.model.layers.17.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
|
| 130 |
+
"language_model.model.layers.17.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
|
| 131 |
+
"language_model.model.layers.17.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
|
| 132 |
+
"language_model.model.layers.17.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
|
| 133 |
+
"language_model.model.layers.17.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
|
| 134 |
+
"language_model.model.layers.17.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
|
| 135 |
+
"language_model.model.layers.17.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
|
| 136 |
+
"language_model.model.layers.17.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
|
| 137 |
+
"language_model.model.layers.17.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 138 |
+
"language_model.model.layers.17.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 139 |
+
"language_model.model.layers.17.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 140 |
+
"language_model.model.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 141 |
+
"language_model.model.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 142 |
+
"language_model.model.layers.18.linear_attn.A_log": "model-00002-of-00002.safetensors",
|
| 143 |
+
"language_model.model.layers.18.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
|
| 144 |
+
"language_model.model.layers.18.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
|
| 145 |
+
"language_model.model.layers.18.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
|
| 146 |
+
"language_model.model.layers.18.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
|
| 147 |
+
"language_model.model.layers.18.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
|
| 148 |
+
"language_model.model.layers.18.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
|
| 149 |
+
"language_model.model.layers.18.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
|
| 150 |
+
"language_model.model.layers.18.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
|
| 151 |
+
"language_model.model.layers.18.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 152 |
+
"language_model.model.layers.18.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 153 |
+
"language_model.model.layers.18.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 154 |
+
"language_model.model.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 155 |
+
"language_model.model.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 156 |
+
"language_model.model.layers.19.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 157 |
+
"language_model.model.layers.19.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 158 |
+
"language_model.model.layers.19.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 159 |
+
"language_model.model.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 160 |
+
"language_model.model.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 161 |
+
"language_model.model.layers.19.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 162 |
+
"language_model.model.layers.19.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 163 |
+
"language_model.model.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 164 |
+
"language_model.model.layers.19.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 165 |
+
"language_model.model.layers.19.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 166 |
+
"language_model.model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 167 |
+
"language_model.model.layers.2.linear_attn.A_log": "model-00001-of-00002.safetensors",
|
| 168 |
+
"language_model.model.layers.2.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
|
| 169 |
+
"language_model.model.layers.2.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
|
| 170 |
+
"language_model.model.layers.2.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
|
| 171 |
+
"language_model.model.layers.2.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
|
| 172 |
+
"language_model.model.layers.2.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
|
| 173 |
+
"language_model.model.layers.2.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
|
| 174 |
+
"language_model.model.layers.2.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
|
| 175 |
+
"language_model.model.layers.2.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
|
| 176 |
+
"language_model.model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 177 |
+
"language_model.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 178 |
+
"language_model.model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 179 |
+
"language_model.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 180 |
+
"language_model.model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 181 |
+
"language_model.model.layers.20.linear_attn.A_log": "model-00002-of-00002.safetensors",
|
| 182 |
+
"language_model.model.layers.20.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
|
| 183 |
+
"language_model.model.layers.20.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
|
| 184 |
+
"language_model.model.layers.20.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
|
| 185 |
+
"language_model.model.layers.20.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
|
| 186 |
+
"language_model.model.layers.20.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
|
| 187 |
+
"language_model.model.layers.20.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
|
| 188 |
+
"language_model.model.layers.20.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
|
| 189 |
+
"language_model.model.layers.20.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
|
| 190 |
+
"language_model.model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 191 |
+
"language_model.model.layers.20.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 192 |
+
"language_model.model.layers.20.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 193 |
+
"language_model.model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 194 |
+
"language_model.model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 195 |
+
"language_model.model.layers.21.linear_attn.A_log": "model-00002-of-00002.safetensors",
|
| 196 |
+
"language_model.model.layers.21.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
|
| 197 |
+
"language_model.model.layers.21.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
|
| 198 |
+
"language_model.model.layers.21.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
|
| 199 |
+
"language_model.model.layers.21.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
|
| 200 |
+
"language_model.model.layers.21.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
|
| 201 |
+
"language_model.model.layers.21.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
|
| 202 |
+
"language_model.model.layers.21.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
|
| 203 |
+
"language_model.model.layers.21.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
|
| 204 |
+
"language_model.model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 205 |
+
"language_model.model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 206 |
+
"language_model.model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 207 |
+
"language_model.model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 208 |
+
"language_model.model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 209 |
+
"language_model.model.layers.22.linear_attn.A_log": "model-00002-of-00002.safetensors",
|
| 210 |
+
"language_model.model.layers.22.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
|
| 211 |
+
"language_model.model.layers.22.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
|
| 212 |
+
"language_model.model.layers.22.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
|
| 213 |
+
"language_model.model.layers.22.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
|
| 214 |
+
"language_model.model.layers.22.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
|
| 215 |
+
"language_model.model.layers.22.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
|
| 216 |
+
"language_model.model.layers.22.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
|
| 217 |
+
"language_model.model.layers.22.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
|
| 218 |
+
"language_model.model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 219 |
+
"language_model.model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 220 |
+
"language_model.model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 221 |
+
"language_model.model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 222 |
+
"language_model.model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 223 |
+
"language_model.model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 224 |
+
"language_model.model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 225 |
+
"language_model.model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 226 |
+
"language_model.model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 227 |
+
"language_model.model.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 228 |
+
"language_model.model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 229 |
+
"language_model.model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 230 |
+
"language_model.model.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 231 |
+
"language_model.model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 232 |
+
"language_model.model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 233 |
+
"language_model.model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 234 |
+
"language_model.model.layers.24.linear_attn.A_log": "model-00002-of-00002.safetensors",
|
| 235 |
+
"language_model.model.layers.24.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
|
| 236 |
+
"language_model.model.layers.24.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
|
| 237 |
+
"language_model.model.layers.24.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
|
| 238 |
+
"language_model.model.layers.24.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
|
| 239 |
+
"language_model.model.layers.24.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
|
| 240 |
+
"language_model.model.layers.24.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
|
| 241 |
+
"language_model.model.layers.24.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
|
| 242 |
+
"language_model.model.layers.24.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
|
| 243 |
+
"language_model.model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 244 |
+
"language_model.model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 245 |
+
"language_model.model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 246 |
+
"language_model.model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 247 |
+
"language_model.model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 248 |
+
"language_model.model.layers.25.linear_attn.A_log": "model-00002-of-00002.safetensors",
|
| 249 |
+
"language_model.model.layers.25.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
|
| 250 |
+
"language_model.model.layers.25.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
|
| 251 |
+
"language_model.model.layers.25.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
|
| 252 |
+
"language_model.model.layers.25.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
|
| 253 |
+
"language_model.model.layers.25.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
|
| 254 |
+
"language_model.model.layers.25.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
|
| 255 |
+
"language_model.model.layers.25.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
|
| 256 |
+
"language_model.model.layers.25.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
|
| 257 |
+
"language_model.model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 258 |
+
"language_model.model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 259 |
+
"language_model.model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 260 |
+
"language_model.model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 261 |
+
"language_model.model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 262 |
+
"language_model.model.layers.26.linear_attn.A_log": "model-00002-of-00002.safetensors",
|
| 263 |
+
"language_model.model.layers.26.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
|
| 264 |
+
"language_model.model.layers.26.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
|
| 265 |
+
"language_model.model.layers.26.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
|
| 266 |
+
"language_model.model.layers.26.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
|
| 267 |
+
"language_model.model.layers.26.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
|
| 268 |
+
"language_model.model.layers.26.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
|
| 269 |
+
"language_model.model.layers.26.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
|
| 270 |
+
"language_model.model.layers.26.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
|
| 271 |
+
"language_model.model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 272 |
+
"language_model.model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 273 |
+
"language_model.model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 274 |
+
"language_model.model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 275 |
+
"language_model.model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 276 |
+
"language_model.model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 277 |
+
"language_model.model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 278 |
+
"language_model.model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 279 |
+
"language_model.model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 280 |
+
"language_model.model.layers.27.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 281 |
+
"language_model.model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 282 |
+
"language_model.model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 283 |
+
"language_model.model.layers.27.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 284 |
+
"language_model.model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 285 |
+
"language_model.model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 286 |
+
"language_model.model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 287 |
+
"language_model.model.layers.28.linear_attn.A_log": "model-00002-of-00002.safetensors",
|
| 288 |
+
"language_model.model.layers.28.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
|
| 289 |
+
"language_model.model.layers.28.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
|
| 290 |
+
"language_model.model.layers.28.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
|
| 291 |
+
"language_model.model.layers.28.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
|
| 292 |
+
"language_model.model.layers.28.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
|
| 293 |
+
"language_model.model.layers.28.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
|
| 294 |
+
"language_model.model.layers.28.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
|
| 295 |
+
"language_model.model.layers.28.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
|
| 296 |
+
"language_model.model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 297 |
+
"language_model.model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 298 |
+
"language_model.model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 299 |
+
"language_model.model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 300 |
+
"language_model.model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 301 |
+
"language_model.model.layers.29.linear_attn.A_log": "model-00002-of-00002.safetensors",
|
| 302 |
+
"language_model.model.layers.29.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
|
| 303 |
+
"language_model.model.layers.29.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
|
| 304 |
+
"language_model.model.layers.29.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
|
| 305 |
+
"language_model.model.layers.29.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
|
| 306 |
+
"language_model.model.layers.29.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
|
| 307 |
+
"language_model.model.layers.29.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
|
| 308 |
+
"language_model.model.layers.29.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
|
| 309 |
+
"language_model.model.layers.29.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
|
| 310 |
+
"language_model.model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 311 |
+
"language_model.model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 312 |
+
"language_model.model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 313 |
+
"language_model.model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 314 |
+
"language_model.model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 315 |
+
"language_model.model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 316 |
+
"language_model.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 317 |
+
"language_model.model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 318 |
+
"language_model.model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 319 |
+
"language_model.model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 320 |
+
"language_model.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 321 |
+
"language_model.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 322 |
+
"language_model.model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 323 |
+
"language_model.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 324 |
+
"language_model.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 325 |
+
"language_model.model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 326 |
+
"language_model.model.layers.30.linear_attn.A_log": "model-00002-of-00002.safetensors",
|
| 327 |
+
"language_model.model.layers.30.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
|
| 328 |
+
"language_model.model.layers.30.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
|
| 329 |
+
"language_model.model.layers.30.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
|
| 330 |
+
"language_model.model.layers.30.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
|
| 331 |
+
"language_model.model.layers.30.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
|
| 332 |
+
"language_model.model.layers.30.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
|
| 333 |
+
"language_model.model.layers.30.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
|
| 334 |
+
"language_model.model.layers.30.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
|
| 335 |
+
"language_model.model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 336 |
+
"language_model.model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 337 |
+
"language_model.model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 338 |
+
"language_model.model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 339 |
+
"language_model.model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 340 |
+
"language_model.model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 341 |
+
"language_model.model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 342 |
+
"language_model.model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 343 |
+
"language_model.model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 344 |
+
"language_model.model.layers.31.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 345 |
+
"language_model.model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 346 |
+
"language_model.model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 347 |
+
"language_model.model.layers.31.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 348 |
+
"language_model.model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 349 |
+
"language_model.model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 350 |
+
"language_model.model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 351 |
+
"language_model.model.layers.4.linear_attn.A_log": "model-00001-of-00002.safetensors",
|
| 352 |
+
"language_model.model.layers.4.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
|
| 353 |
+
"language_model.model.layers.4.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
|
| 354 |
+
"language_model.model.layers.4.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
|
| 355 |
+
"language_model.model.layers.4.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
|
| 356 |
+
"language_model.model.layers.4.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
|
| 357 |
+
"language_model.model.layers.4.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
|
| 358 |
+
"language_model.model.layers.4.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
|
| 359 |
+
"language_model.model.layers.4.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
|
| 360 |
+
"language_model.model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 361 |
+
"language_model.model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 362 |
+
"language_model.model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 363 |
+
"language_model.model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 364 |
+
"language_model.model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 365 |
+
"language_model.model.layers.5.linear_attn.A_log": "model-00001-of-00002.safetensors",
|
| 366 |
+
"language_model.model.layers.5.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
|
| 367 |
+
"language_model.model.layers.5.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
|
| 368 |
+
"language_model.model.layers.5.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
|
| 369 |
+
"language_model.model.layers.5.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
|
| 370 |
+
"language_model.model.layers.5.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
|
| 371 |
+
"language_model.model.layers.5.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
|
| 372 |
+
"language_model.model.layers.5.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
|
| 373 |
+
"language_model.model.layers.5.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
|
| 374 |
+
"language_model.model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 375 |
+
"language_model.model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 376 |
+
"language_model.model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 377 |
+
"language_model.model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 378 |
+
"language_model.model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 379 |
+
"language_model.model.layers.6.linear_attn.A_log": "model-00001-of-00002.safetensors",
|
| 380 |
+
"language_model.model.layers.6.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
|
| 381 |
+
"language_model.model.layers.6.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
|
| 382 |
+
"language_model.model.layers.6.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
|
| 383 |
+
"language_model.model.layers.6.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
|
| 384 |
+
"language_model.model.layers.6.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
|
| 385 |
+
"language_model.model.layers.6.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
|
| 386 |
+
"language_model.model.layers.6.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
|
| 387 |
+
"language_model.model.layers.6.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
|
| 388 |
+
"language_model.model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 389 |
+
"language_model.model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 390 |
+
"language_model.model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 391 |
+
"language_model.model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 392 |
+
"language_model.model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 393 |
+
"language_model.model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 394 |
+
"language_model.model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 395 |
+
"language_model.model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 396 |
+
"language_model.model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 397 |
+
"language_model.model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 398 |
+
"language_model.model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 399 |
+
"language_model.model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 400 |
+
"language_model.model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 401 |
+
"language_model.model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 402 |
+
"language_model.model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 403 |
+
"language_model.model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 404 |
+
"language_model.model.layers.8.linear_attn.A_log": "model-00001-of-00002.safetensors",
|
| 405 |
+
"language_model.model.layers.8.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
|
| 406 |
+
"language_model.model.layers.8.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
|
| 407 |
+
"language_model.model.layers.8.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
|
| 408 |
+
"language_model.model.layers.8.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
|
| 409 |
+
"language_model.model.layers.8.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
|
| 410 |
+
"language_model.model.layers.8.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
|
| 411 |
+
"language_model.model.layers.8.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
|
| 412 |
+
"language_model.model.layers.8.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
|
| 413 |
+
"language_model.model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 414 |
+
"language_model.model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 415 |
+
"language_model.model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 416 |
+
"language_model.model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 417 |
+
"language_model.model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 418 |
+
"language_model.model.layers.9.linear_attn.A_log": "model-00001-of-00002.safetensors",
|
| 419 |
+
"language_model.model.layers.9.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
|
| 420 |
+
"language_model.model.layers.9.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
|
| 421 |
+
"language_model.model.layers.9.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
|
| 422 |
+
"language_model.model.layers.9.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
|
| 423 |
+
"language_model.model.layers.9.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
|
| 424 |
+
"language_model.model.layers.9.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
|
| 425 |
+
"language_model.model.layers.9.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
|
| 426 |
+
"language_model.model.layers.9.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
|
| 427 |
+
"language_model.model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 428 |
+
"language_model.model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 429 |
+
"language_model.model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 430 |
+
"language_model.model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 431 |
+
"language_model.model.norm.weight": "model-00002-of-00002.safetensors",
|
| 432 |
+
"vision_tower.blocks.0.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 433 |
+
"vision_tower.blocks.0.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 434 |
+
"vision_tower.blocks.0.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 435 |
+
"vision_tower.blocks.0.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 436 |
+
"vision_tower.blocks.0.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 437 |
+
"vision_tower.blocks.0.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 438 |
+
"vision_tower.blocks.0.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 439 |
+
"vision_tower.blocks.0.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 440 |
+
"vision_tower.blocks.0.norm1.bias": "model-00001-of-00002.safetensors",
|
| 441 |
+
"vision_tower.blocks.0.norm1.weight": "model-00001-of-00002.safetensors",
|
| 442 |
+
"vision_tower.blocks.0.norm2.bias": "model-00001-of-00002.safetensors",
|
| 443 |
+
"vision_tower.blocks.0.norm2.weight": "model-00001-of-00002.safetensors",
|
| 444 |
+
"vision_tower.blocks.1.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 445 |
+
"vision_tower.blocks.1.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 446 |
+
"vision_tower.blocks.1.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 447 |
+
"vision_tower.blocks.1.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 448 |
+
"vision_tower.blocks.1.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 449 |
+
"vision_tower.blocks.1.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 450 |
+
"vision_tower.blocks.1.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 451 |
+
"vision_tower.blocks.1.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 452 |
+
"vision_tower.blocks.1.norm1.bias": "model-00001-of-00002.safetensors",
|
| 453 |
+
"vision_tower.blocks.1.norm1.weight": "model-00001-of-00002.safetensors",
|
| 454 |
+
"vision_tower.blocks.1.norm2.bias": "model-00001-of-00002.safetensors",
|
| 455 |
+
"vision_tower.blocks.1.norm2.weight": "model-00001-of-00002.safetensors",
|
| 456 |
+
"vision_tower.blocks.10.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 457 |
+
"vision_tower.blocks.10.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 458 |
+
"vision_tower.blocks.10.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 459 |
+
"vision_tower.blocks.10.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 460 |
+
"vision_tower.blocks.10.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 461 |
+
"vision_tower.blocks.10.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 462 |
+
"vision_tower.blocks.10.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 463 |
+
"vision_tower.blocks.10.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 464 |
+
"vision_tower.blocks.10.norm1.bias": "model-00001-of-00002.safetensors",
|
| 465 |
+
"vision_tower.blocks.10.norm1.weight": "model-00001-of-00002.safetensors",
|
| 466 |
+
"vision_tower.blocks.10.norm2.bias": "model-00001-of-00002.safetensors",
|
| 467 |
+
"vision_tower.blocks.10.norm2.weight": "model-00001-of-00002.safetensors",
|
| 468 |
+
"vision_tower.blocks.11.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 469 |
+
"vision_tower.blocks.11.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 470 |
+
"vision_tower.blocks.11.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 471 |
+
"vision_tower.blocks.11.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 472 |
+
"vision_tower.blocks.11.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 473 |
+
"vision_tower.blocks.11.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 474 |
+
"vision_tower.blocks.11.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 475 |
+
"vision_tower.blocks.11.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 476 |
+
"vision_tower.blocks.11.norm1.bias": "model-00001-of-00002.safetensors",
|
| 477 |
+
"vision_tower.blocks.11.norm1.weight": "model-00001-of-00002.safetensors",
|
| 478 |
+
"vision_tower.blocks.11.norm2.bias": "model-00001-of-00002.safetensors",
|
| 479 |
+
"vision_tower.blocks.11.norm2.weight": "model-00001-of-00002.safetensors",
|
| 480 |
+
"vision_tower.blocks.12.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 481 |
+
"vision_tower.blocks.12.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 482 |
+
"vision_tower.blocks.12.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 483 |
+
"vision_tower.blocks.12.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 484 |
+
"vision_tower.blocks.12.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 485 |
+
"vision_tower.blocks.12.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 486 |
+
"vision_tower.blocks.12.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 487 |
+
"vision_tower.blocks.12.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 488 |
+
"vision_tower.blocks.12.norm1.bias": "model-00001-of-00002.safetensors",
|
| 489 |
+
"vision_tower.blocks.12.norm1.weight": "model-00001-of-00002.safetensors",
|
| 490 |
+
"vision_tower.blocks.12.norm2.bias": "model-00001-of-00002.safetensors",
|
| 491 |
+
"vision_tower.blocks.12.norm2.weight": "model-00001-of-00002.safetensors",
|
| 492 |
+
"vision_tower.blocks.13.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 493 |
+
"vision_tower.blocks.13.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 494 |
+
"vision_tower.blocks.13.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 495 |
+
"vision_tower.blocks.13.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 496 |
+
"vision_tower.blocks.13.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 497 |
+
"vision_tower.blocks.13.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 498 |
+
"vision_tower.blocks.13.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 499 |
+
"vision_tower.blocks.13.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 500 |
+
"vision_tower.blocks.13.norm1.bias": "model-00001-of-00002.safetensors",
|
| 501 |
+
"vision_tower.blocks.13.norm1.weight": "model-00001-of-00002.safetensors",
|
| 502 |
+
"vision_tower.blocks.13.norm2.bias": "model-00001-of-00002.safetensors",
|
| 503 |
+
"vision_tower.blocks.13.norm2.weight": "model-00001-of-00002.safetensors",
|
| 504 |
+
"vision_tower.blocks.14.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 505 |
+
"vision_tower.blocks.14.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 506 |
+
"vision_tower.blocks.14.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 507 |
+
"vision_tower.blocks.14.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 508 |
+
"vision_tower.blocks.14.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 509 |
+
"vision_tower.blocks.14.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 510 |
+
"vision_tower.blocks.14.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 511 |
+
"vision_tower.blocks.14.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 512 |
+
"vision_tower.blocks.14.norm1.bias": "model-00001-of-00002.safetensors",
|
| 513 |
+
"vision_tower.blocks.14.norm1.weight": "model-00001-of-00002.safetensors",
|
| 514 |
+
"vision_tower.blocks.14.norm2.bias": "model-00001-of-00002.safetensors",
|
| 515 |
+
"vision_tower.blocks.14.norm2.weight": "model-00001-of-00002.safetensors",
|
| 516 |
+
"vision_tower.blocks.15.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 517 |
+
"vision_tower.blocks.15.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 518 |
+
"vision_tower.blocks.15.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 519 |
+
"vision_tower.blocks.15.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 520 |
+
"vision_tower.blocks.15.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 521 |
+
"vision_tower.blocks.15.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 522 |
+
"vision_tower.blocks.15.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 523 |
+
"vision_tower.blocks.15.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 524 |
+
"vision_tower.blocks.15.norm1.bias": "model-00001-of-00002.safetensors",
|
| 525 |
+
"vision_tower.blocks.15.norm1.weight": "model-00001-of-00002.safetensors",
|
| 526 |
+
"vision_tower.blocks.15.norm2.bias": "model-00001-of-00002.safetensors",
|
| 527 |
+
"vision_tower.blocks.15.norm2.weight": "model-00001-of-00002.safetensors",
|
| 528 |
+
"vision_tower.blocks.16.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 529 |
+
"vision_tower.blocks.16.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 530 |
+
"vision_tower.blocks.16.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 531 |
+
"vision_tower.blocks.16.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 532 |
+
"vision_tower.blocks.16.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 533 |
+
"vision_tower.blocks.16.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 534 |
+
"vision_tower.blocks.16.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 535 |
+
"vision_tower.blocks.16.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 536 |
+
"vision_tower.blocks.16.norm1.bias": "model-00001-of-00002.safetensors",
|
| 537 |
+
"vision_tower.blocks.16.norm1.weight": "model-00001-of-00002.safetensors",
|
| 538 |
+
"vision_tower.blocks.16.norm2.bias": "model-00001-of-00002.safetensors",
|
| 539 |
+
"vision_tower.blocks.16.norm2.weight": "model-00001-of-00002.safetensors",
|
| 540 |
+
"vision_tower.blocks.17.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 541 |
+
"vision_tower.blocks.17.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 542 |
+
"vision_tower.blocks.17.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 543 |
+
"vision_tower.blocks.17.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 544 |
+
"vision_tower.blocks.17.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 545 |
+
"vision_tower.blocks.17.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 546 |
+
"vision_tower.blocks.17.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 547 |
+
"vision_tower.blocks.17.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 548 |
+
"vision_tower.blocks.17.norm1.bias": "model-00001-of-00002.safetensors",
|
| 549 |
+
"vision_tower.blocks.17.norm1.weight": "model-00001-of-00002.safetensors",
|
| 550 |
+
"vision_tower.blocks.17.norm2.bias": "model-00001-of-00002.safetensors",
|
| 551 |
+
"vision_tower.blocks.17.norm2.weight": "model-00001-of-00002.safetensors",
|
| 552 |
+
"vision_tower.blocks.18.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 553 |
+
"vision_tower.blocks.18.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 554 |
+
"vision_tower.blocks.18.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 555 |
+
"vision_tower.blocks.18.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 556 |
+
"vision_tower.blocks.18.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 557 |
+
"vision_tower.blocks.18.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 558 |
+
"vision_tower.blocks.18.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 559 |
+
"vision_tower.blocks.18.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 560 |
+
"vision_tower.blocks.18.norm1.bias": "model-00001-of-00002.safetensors",
|
| 561 |
+
"vision_tower.blocks.18.norm1.weight": "model-00001-of-00002.safetensors",
|
| 562 |
+
"vision_tower.blocks.18.norm2.bias": "model-00001-of-00002.safetensors",
|
| 563 |
+
"vision_tower.blocks.18.norm2.weight": "model-00001-of-00002.safetensors",
|
| 564 |
+
"vision_tower.blocks.19.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 565 |
+
"vision_tower.blocks.19.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 566 |
+
"vision_tower.blocks.19.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 567 |
+
"vision_tower.blocks.19.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 568 |
+
"vision_tower.blocks.19.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 569 |
+
"vision_tower.blocks.19.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 570 |
+
"vision_tower.blocks.19.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 571 |
+
"vision_tower.blocks.19.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 572 |
+
"vision_tower.blocks.19.norm1.bias": "model-00001-of-00002.safetensors",
|
| 573 |
+
"vision_tower.blocks.19.norm1.weight": "model-00001-of-00002.safetensors",
|
| 574 |
+
"vision_tower.blocks.19.norm2.bias": "model-00001-of-00002.safetensors",
|
| 575 |
+
"vision_tower.blocks.19.norm2.weight": "model-00001-of-00002.safetensors",
|
| 576 |
+
"vision_tower.blocks.2.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 577 |
+
"vision_tower.blocks.2.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 578 |
+
"vision_tower.blocks.2.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 579 |
+
"vision_tower.blocks.2.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 580 |
+
"vision_tower.blocks.2.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 581 |
+
"vision_tower.blocks.2.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 582 |
+
"vision_tower.blocks.2.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 583 |
+
"vision_tower.blocks.2.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 584 |
+
"vision_tower.blocks.2.norm1.bias": "model-00001-of-00002.safetensors",
|
| 585 |
+
"vision_tower.blocks.2.norm1.weight": "model-00001-of-00002.safetensors",
|
| 586 |
+
"vision_tower.blocks.2.norm2.bias": "model-00001-of-00002.safetensors",
|
| 587 |
+
"vision_tower.blocks.2.norm2.weight": "model-00001-of-00002.safetensors",
|
| 588 |
+
"vision_tower.blocks.20.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 589 |
+
"vision_tower.blocks.20.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 590 |
+
"vision_tower.blocks.20.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 591 |
+
"vision_tower.blocks.20.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 592 |
+
"vision_tower.blocks.20.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 593 |
+
"vision_tower.blocks.20.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 594 |
+
"vision_tower.blocks.20.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 595 |
+
"vision_tower.blocks.20.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 596 |
+
"vision_tower.blocks.20.norm1.bias": "model-00001-of-00002.safetensors",
|
| 597 |
+
"vision_tower.blocks.20.norm1.weight": "model-00001-of-00002.safetensors",
|
| 598 |
+
"vision_tower.blocks.20.norm2.bias": "model-00001-of-00002.safetensors",
|
| 599 |
+
"vision_tower.blocks.20.norm2.weight": "model-00001-of-00002.safetensors",
|
| 600 |
+
"vision_tower.blocks.21.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 601 |
+
"vision_tower.blocks.21.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 602 |
+
"vision_tower.blocks.21.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 603 |
+
"vision_tower.blocks.21.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 604 |
+
"vision_tower.blocks.21.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 605 |
+
"vision_tower.blocks.21.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 606 |
+
"vision_tower.blocks.21.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 607 |
+
"vision_tower.blocks.21.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 608 |
+
"vision_tower.blocks.21.norm1.bias": "model-00001-of-00002.safetensors",
|
| 609 |
+
"vision_tower.blocks.21.norm1.weight": "model-00001-of-00002.safetensors",
|
| 610 |
+
"vision_tower.blocks.21.norm2.bias": "model-00001-of-00002.safetensors",
|
| 611 |
+
"vision_tower.blocks.21.norm2.weight": "model-00001-of-00002.safetensors",
|
| 612 |
+
"vision_tower.blocks.22.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 613 |
+
"vision_tower.blocks.22.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 614 |
+
"vision_tower.blocks.22.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 615 |
+
"vision_tower.blocks.22.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 616 |
+
"vision_tower.blocks.22.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 617 |
+
"vision_tower.blocks.22.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 618 |
+
"vision_tower.blocks.22.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 619 |
+
"vision_tower.blocks.22.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 620 |
+
"vision_tower.blocks.22.norm1.bias": "model-00001-of-00002.safetensors",
|
| 621 |
+
"vision_tower.blocks.22.norm1.weight": "model-00001-of-00002.safetensors",
|
| 622 |
+
"vision_tower.blocks.22.norm2.bias": "model-00001-of-00002.safetensors",
|
| 623 |
+
"vision_tower.blocks.22.norm2.weight": "model-00001-of-00002.safetensors",
|
| 624 |
+
"vision_tower.blocks.23.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 625 |
+
"vision_tower.blocks.23.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 626 |
+
"vision_tower.blocks.23.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 627 |
+
"vision_tower.blocks.23.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 628 |
+
"vision_tower.blocks.23.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 629 |
+
"vision_tower.blocks.23.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 630 |
+
"vision_tower.blocks.23.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 631 |
+
"vision_tower.blocks.23.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 632 |
+
"vision_tower.blocks.23.norm1.bias": "model-00001-of-00002.safetensors",
|
| 633 |
+
"vision_tower.blocks.23.norm1.weight": "model-00001-of-00002.safetensors",
|
| 634 |
+
"vision_tower.blocks.23.norm2.bias": "model-00001-of-00002.safetensors",
|
| 635 |
+
"vision_tower.blocks.23.norm2.weight": "model-00001-of-00002.safetensors",
|
| 636 |
+
"vision_tower.blocks.3.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 637 |
+
"vision_tower.blocks.3.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 638 |
+
"vision_tower.blocks.3.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 639 |
+
"vision_tower.blocks.3.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 640 |
+
"vision_tower.blocks.3.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 641 |
+
"vision_tower.blocks.3.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 642 |
+
"vision_tower.blocks.3.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 643 |
+
"vision_tower.blocks.3.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 644 |
+
"vision_tower.blocks.3.norm1.bias": "model-00001-of-00002.safetensors",
|
| 645 |
+
"vision_tower.blocks.3.norm1.weight": "model-00001-of-00002.safetensors",
|
| 646 |
+
"vision_tower.blocks.3.norm2.bias": "model-00001-of-00002.safetensors",
|
| 647 |
+
"vision_tower.blocks.3.norm2.weight": "model-00001-of-00002.safetensors",
|
| 648 |
+
"vision_tower.blocks.4.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 649 |
+
"vision_tower.blocks.4.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 650 |
+
"vision_tower.blocks.4.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 651 |
+
"vision_tower.blocks.4.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 652 |
+
"vision_tower.blocks.4.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 653 |
+
"vision_tower.blocks.4.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 654 |
+
"vision_tower.blocks.4.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 655 |
+
"vision_tower.blocks.4.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 656 |
+
"vision_tower.blocks.4.norm1.bias": "model-00001-of-00002.safetensors",
|
| 657 |
+
"vision_tower.blocks.4.norm1.weight": "model-00001-of-00002.safetensors",
|
| 658 |
+
"vision_tower.blocks.4.norm2.bias": "model-00001-of-00002.safetensors",
|
| 659 |
+
"vision_tower.blocks.4.norm2.weight": "model-00001-of-00002.safetensors",
|
| 660 |
+
"vision_tower.blocks.5.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 661 |
+
"vision_tower.blocks.5.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 662 |
+
"vision_tower.blocks.5.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 663 |
+
"vision_tower.blocks.5.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 664 |
+
"vision_tower.blocks.5.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 665 |
+
"vision_tower.blocks.5.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 666 |
+
"vision_tower.blocks.5.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 667 |
+
"vision_tower.blocks.5.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 668 |
+
"vision_tower.blocks.5.norm1.bias": "model-00001-of-00002.safetensors",
|
| 669 |
+
"vision_tower.blocks.5.norm1.weight": "model-00001-of-00002.safetensors",
|
| 670 |
+
"vision_tower.blocks.5.norm2.bias": "model-00001-of-00002.safetensors",
|
| 671 |
+
"vision_tower.blocks.5.norm2.weight": "model-00001-of-00002.safetensors",
|
| 672 |
+
"vision_tower.blocks.6.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 673 |
+
"vision_tower.blocks.6.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 674 |
+
"vision_tower.blocks.6.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 675 |
+
"vision_tower.blocks.6.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 676 |
+
"vision_tower.blocks.6.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 677 |
+
"vision_tower.blocks.6.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 678 |
+
"vision_tower.blocks.6.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 679 |
+
"vision_tower.blocks.6.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 680 |
+
"vision_tower.blocks.6.norm1.bias": "model-00001-of-00002.safetensors",
|
| 681 |
+
"vision_tower.blocks.6.norm1.weight": "model-00001-of-00002.safetensors",
|
| 682 |
+
"vision_tower.blocks.6.norm2.bias": "model-00001-of-00002.safetensors",
|
| 683 |
+
"vision_tower.blocks.6.norm2.weight": "model-00001-of-00002.safetensors",
|
| 684 |
+
"vision_tower.blocks.7.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 685 |
+
"vision_tower.blocks.7.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 686 |
+
"vision_tower.blocks.7.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 687 |
+
"vision_tower.blocks.7.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 688 |
+
"vision_tower.blocks.7.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 689 |
+
"vision_tower.blocks.7.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 690 |
+
"vision_tower.blocks.7.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 691 |
+
"vision_tower.blocks.7.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 692 |
+
"vision_tower.blocks.7.norm1.bias": "model-00001-of-00002.safetensors",
|
| 693 |
+
"vision_tower.blocks.7.norm1.weight": "model-00001-of-00002.safetensors",
|
| 694 |
+
"vision_tower.blocks.7.norm2.bias": "model-00001-of-00002.safetensors",
|
| 695 |
+
"vision_tower.blocks.7.norm2.weight": "model-00001-of-00002.safetensors",
|
| 696 |
+
"vision_tower.blocks.8.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 697 |
+
"vision_tower.blocks.8.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 698 |
+
"vision_tower.blocks.8.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 699 |
+
"vision_tower.blocks.8.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 700 |
+
"vision_tower.blocks.8.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 701 |
+
"vision_tower.blocks.8.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 702 |
+
"vision_tower.blocks.8.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 703 |
+
"vision_tower.blocks.8.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 704 |
+
"vision_tower.blocks.8.norm1.bias": "model-00001-of-00002.safetensors",
|
| 705 |
+
"vision_tower.blocks.8.norm1.weight": "model-00001-of-00002.safetensors",
|
| 706 |
+
"vision_tower.blocks.8.norm2.bias": "model-00001-of-00002.safetensors",
|
| 707 |
+
"vision_tower.blocks.8.norm2.weight": "model-00001-of-00002.safetensors",
|
| 708 |
+
"vision_tower.blocks.9.attn.proj.bias": "model-00001-of-00002.safetensors",
|
| 709 |
+
"vision_tower.blocks.9.attn.proj.weight": "model-00001-of-00002.safetensors",
|
| 710 |
+
"vision_tower.blocks.9.attn.qkv.bias": "model-00001-of-00002.safetensors",
|
| 711 |
+
"vision_tower.blocks.9.attn.qkv.weight": "model-00001-of-00002.safetensors",
|
| 712 |
+
"vision_tower.blocks.9.mlp.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 713 |
+
"vision_tower.blocks.9.mlp.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 714 |
+
"vision_tower.blocks.9.mlp.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 715 |
+
"vision_tower.blocks.9.mlp.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 716 |
+
"vision_tower.blocks.9.norm1.bias": "model-00001-of-00002.safetensors",
|
| 717 |
+
"vision_tower.blocks.9.norm1.weight": "model-00001-of-00002.safetensors",
|
| 718 |
+
"vision_tower.blocks.9.norm2.bias": "model-00001-of-00002.safetensors",
|
| 719 |
+
"vision_tower.blocks.9.norm2.weight": "model-00001-of-00002.safetensors",
|
| 720 |
+
"vision_tower.merger.linear_fc1.bias": "model-00001-of-00002.safetensors",
|
| 721 |
+
"vision_tower.merger.linear_fc1.weight": "model-00001-of-00002.safetensors",
|
| 722 |
+
"vision_tower.merger.linear_fc2.bias": "model-00001-of-00002.safetensors",
|
| 723 |
+
"vision_tower.merger.linear_fc2.weight": "model-00001-of-00002.safetensors",
|
| 724 |
+
"vision_tower.merger.norm.bias": "model-00001-of-00002.safetensors",
|
| 725 |
+
"vision_tower.merger.norm.weight": "model-00001-of-00002.safetensors",
|
| 726 |
+
"vision_tower.patch_embed.proj.bias": "model-00001-of-00002.safetensors",
|
| 727 |
+
"vision_tower.patch_embed.proj.weight": "model-00001-of-00002.safetensors",
|
| 728 |
+
"vision_tower.pos_embed.weight": "model-00001-of-00002.safetensors"
|
| 729 |
+
}
|
| 730 |
+
}
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"size": {
|
| 3 |
+
"longest_edge": 16777216,
|
| 4 |
+
"shortest_edge": 65536
|
| 5 |
+
},
|
| 6 |
+
"patch_size": 16,
|
| 7 |
+
"temporal_patch_size": 2,
|
| 8 |
+
"merge_size": 2,
|
| 9 |
+
"image_mean": [
|
| 10 |
+
0.5,
|
| 11 |
+
0.5,
|
| 12 |
+
0.5
|
| 13 |
+
],
|
| 14 |
+
"image_std": [
|
| 15 |
+
0.5,
|
| 16 |
+
0.5,
|
| 17 |
+
0.5
|
| 18 |
+
],
|
| 19 |
+
"processor_class": "Qwen3VLProcessor",
|
| 20 |
+
"image_processor_type": "Qwen2VLImageProcessorFast"
|
| 21 |
+
}
|
processor_config.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"image_processor": {
|
| 3 |
+
"data_format": "channels_first",
|
| 4 |
+
"do_convert_rgb": true,
|
| 5 |
+
"do_normalize": true,
|
| 6 |
+
"do_rescale": true,
|
| 7 |
+
"do_resize": true,
|
| 8 |
+
"image_mean": [
|
| 9 |
+
0.5,
|
| 10 |
+
0.5,
|
| 11 |
+
0.5
|
| 12 |
+
],
|
| 13 |
+
"image_processor_type": "Qwen2VLImageProcessorFast",
|
| 14 |
+
"image_std": [
|
| 15 |
+
0.5,
|
| 16 |
+
0.5,
|
| 17 |
+
0.5
|
| 18 |
+
],
|
| 19 |
+
"merge_size": 2,
|
| 20 |
+
"patch_size": 16,
|
| 21 |
+
"resample": 3,
|
| 22 |
+
"rescale_factor": 0.00392156862745098,
|
| 23 |
+
"size": {
|
| 24 |
+
"longest_edge": 16777216,
|
| 25 |
+
"shortest_edge": 65536
|
| 26 |
+
},
|
| 27 |
+
"temporal_patch_size": 2
|
| 28 |
+
},
|
| 29 |
+
"processor_class": "Qwen3VLProcessor",
|
| 30 |
+
"video_processor": {
|
| 31 |
+
"data_format": "channels_first",
|
| 32 |
+
"default_to_square": true,
|
| 33 |
+
"do_convert_rgb": true,
|
| 34 |
+
"do_normalize": true,
|
| 35 |
+
"do_rescale": true,
|
| 36 |
+
"do_resize": true,
|
| 37 |
+
"do_sample_frames": true,
|
| 38 |
+
"fps": 2,
|
| 39 |
+
"image_mean": [
|
| 40 |
+
0.5,
|
| 41 |
+
0.5,
|
| 42 |
+
0.5
|
| 43 |
+
],
|
| 44 |
+
"image_std": [
|
| 45 |
+
0.5,
|
| 46 |
+
0.5,
|
| 47 |
+
0.5
|
| 48 |
+
],
|
| 49 |
+
"max_frames": 768,
|
| 50 |
+
"merge_size": 2,
|
| 51 |
+
"min_frames": 4,
|
| 52 |
+
"patch_size": 16,
|
| 53 |
+
"resample": 3,
|
| 54 |
+
"rescale_factor": 0.00392156862745098,
|
| 55 |
+
"return_metadata": false,
|
| 56 |
+
"size": {
|
| 57 |
+
"longest_edge": 25165824,
|
| 58 |
+
"shortest_edge": 4096
|
| 59 |
+
},
|
| 60 |
+
"temporal_patch_size": 2,
|
| 61 |
+
"video_processor_type": "Qwen3VLVideoProcessor"
|
| 62 |
+
}
|
| 63 |
+
}
|
temperatures.json
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"dev": "experiments/decision/dev-v2-lora-v2b.jsonl",
|
| 3 |
+
"eval": "experiments/decision/eval-v2-lora-v2b.jsonl",
|
| 4 |
+
"questions": {
|
| 5 |
+
"grounding": {
|
| 6 |
+
"temperature": 1.2,
|
| 7 |
+
"dev_items": 771,
|
| 8 |
+
"dev_ece": {
|
| 9 |
+
"before": 0.1964,
|
| 10 |
+
"after": 0.145
|
| 11 |
+
},
|
| 12 |
+
"dev_log_loss": {
|
| 13 |
+
"before": 1.2577,
|
| 14 |
+
"after": 1.2158
|
| 15 |
+
},
|
| 16 |
+
"eval_items": 2079,
|
| 17 |
+
"eval_ece": {
|
| 18 |
+
"before": 0.1447,
|
| 19 |
+
"after": 0.1069
|
| 20 |
+
},
|
| 21 |
+
"eval_log_loss": {
|
| 22 |
+
"before": 1.2496,
|
| 23 |
+
"after": 1.2284
|
| 24 |
+
}
|
| 25 |
+
},
|
| 26 |
+
"action": {
|
| 27 |
+
"temperature": 2.05,
|
| 28 |
+
"dev_items": 1317,
|
| 29 |
+
"dev_ece": {
|
| 30 |
+
"before": 0.2948,
|
| 31 |
+
"after": 0.0523
|
| 32 |
+
},
|
| 33 |
+
"dev_log_loss": {
|
| 34 |
+
"before": 2.1282,
|
| 35 |
+
"after": 1.6834
|
| 36 |
+
},
|
| 37 |
+
"eval_items": 3442,
|
| 38 |
+
"eval_ece": {
|
| 39 |
+
"before": 0.2593,
|
| 40 |
+
"after": 0.0467
|
| 41 |
+
},
|
| 42 |
+
"eval_log_loss": {
|
| 43 |
+
"before": 2.0045,
|
| 44 |
+
"after": 1.6253
|
| 45 |
+
}
|
| 46 |
+
},
|
| 47 |
+
"next_target": {
|
| 48 |
+
"temperature": 0.9,
|
| 49 |
+
"dev_items": 771,
|
| 50 |
+
"dev_ece": {
|
| 51 |
+
"before": 0.1169,
|
| 52 |
+
"after": 0.1149
|
| 53 |
+
},
|
| 54 |
+
"dev_log_loss": {
|
| 55 |
+
"before": 2.9552,
|
| 56 |
+
"after": 2.9388
|
| 57 |
+
},
|
| 58 |
+
"eval_items": 2079,
|
| 59 |
+
"eval_ece": {
|
| 60 |
+
"before": 0.0635,
|
| 61 |
+
"after": 0.0852
|
| 62 |
+
},
|
| 63 |
+
"eval_log_loss": {
|
| 64 |
+
"before": 3.1757,
|
| 65 |
+
"after": 3.1547
|
| 66 |
+
}
|
| 67 |
+
}
|
| 68 |
+
}
|
| 69 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87a7830d63fcf43bf241c3c5242e96e62dd3fdc29224ca26fed8ea333db72de4
|
| 3 |
+
size 19989343
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"audio_bos_token": "<|audio_start|>",
|
| 4 |
+
"audio_eos_token": "<|audio_end|>",
|
| 5 |
+
"audio_token": "<|audio_pad|>",
|
| 6 |
+
"backend": "tokenizers",
|
| 7 |
+
"bos_token": null,
|
| 8 |
+
"clean_up_tokenization_spaces": false,
|
| 9 |
+
"eos_token": "<|im_end|>",
|
| 10 |
+
"errors": "replace",
|
| 11 |
+
"image_token": "<|image_pad|>",
|
| 12 |
+
"is_local": true,
|
| 13 |
+
"model_max_length": 262144,
|
| 14 |
+
"model_specific_special_tokens": {
|
| 15 |
+
"audio_bos_token": "<|audio_start|>",
|
| 16 |
+
"audio_eos_token": "<|audio_end|>",
|
| 17 |
+
"audio_token": "<|audio_pad|>",
|
| 18 |
+
"image_token": "<|image_pad|>",
|
| 19 |
+
"video_token": "<|video_pad|>",
|
| 20 |
+
"vision_bos_token": "<|vision_start|>",
|
| 21 |
+
"vision_eos_token": "<|vision_end|>"
|
| 22 |
+
},
|
| 23 |
+
"pad_token": "<|endoftext|>",
|
| 24 |
+
"pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
|
| 25 |
+
"processor_class": "Qwen3VLProcessor",
|
| 26 |
+
"split_special_tokens": false,
|
| 27 |
+
"tokenizer_class": "TokenizersBackend",
|
| 28 |
+
"unk_token": null,
|
| 29 |
+
"video_token": "<|video_pad|>",
|
| 30 |
+
"vision_bos_token": "<|vision_start|>",
|
| 31 |
+
"vision_eos_token": "<|vision_end|>"
|
| 32 |
+
}
|
video_preprocessor_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"size": {
|
| 3 |
+
"longest_edge": 25165824,
|
| 4 |
+
"shortest_edge": 4096
|
| 5 |
+
},
|
| 6 |
+
"patch_size": 16,
|
| 7 |
+
"temporal_patch_size": 2,
|
| 8 |
+
"merge_size": 2,
|
| 9 |
+
"image_mean": [
|
| 10 |
+
0.5,
|
| 11 |
+
0.5,
|
| 12 |
+
0.5
|
| 13 |
+
],
|
| 14 |
+
"image_std": [
|
| 15 |
+
0.5,
|
| 16 |
+
0.5,
|
| 17 |
+
0.5
|
| 18 |
+
],
|
| 19 |
+
"processor_class": "Qwen3VLProcessor",
|
| 20 |
+
"video_processor_type": "Qwen3VLVideoProcessor"
|
| 21 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|