SenseNova-Loop-9B-Preview (9937e91) MLX 4-bit: weights, MLX model code, card, Apache-2.0 LICENSE and NOTICE
Browse files- .gitattributes +1 -0
- LICENSE +202 -0
- NOTICE +11 -0
- README.md +70 -0
- chat_template.jinja +154 -0
- config.json +134 -0
- generation_config.json +12 -0
- model.safetensors +3 -0
- model.safetensors.index.json +1031 -0
- sensenova_loop.py +259 -0
- tokenizer.json +3 -0
- tokenizer_config.json +34 -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 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
Apache License
|
| 3 |
+
Version 2.0, January 2004
|
| 4 |
+
http://www.apache.org/licenses/
|
| 5 |
+
|
| 6 |
+
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
|
| 7 |
+
|
| 8 |
+
1. Definitions.
|
| 9 |
+
|
| 10 |
+
"License" shall mean the terms and conditions for use, reproduction,
|
| 11 |
+
and distribution as defined by Sections 1 through 9 of this document.
|
| 12 |
+
|
| 13 |
+
"Licensor" shall mean the copyright owner or entity authorized by
|
| 14 |
+
the copyright owner that is granting the License.
|
| 15 |
+
|
| 16 |
+
"Legal Entity" shall mean the union of the acting entity and all
|
| 17 |
+
other entities that control, are controlled by, or are under common
|
| 18 |
+
control with that entity. For the purposes of this definition,
|
| 19 |
+
"control" means (i) the power, direct or indirect, to cause the
|
| 20 |
+
direction or management of such entity, whether by contract or
|
| 21 |
+
otherwise, or (ii) ownership of fifty percent (50%) or more of the
|
| 22 |
+
outstanding shares, or (iii) beneficial ownership of such entity.
|
| 23 |
+
|
| 24 |
+
"You" (or "Your") shall mean an individual or Legal Entity
|
| 25 |
+
exercising permissions granted by this License.
|
| 26 |
+
|
| 27 |
+
"Source" form shall mean the preferred form for making modifications,
|
| 28 |
+
including but not limited to software source code, documentation
|
| 29 |
+
source, and configuration files.
|
| 30 |
+
|
| 31 |
+
"Object" form shall mean any form resulting from mechanical
|
| 32 |
+
transformation or translation of a Source form, including but
|
| 33 |
+
not limited to compiled object code, generated documentation,
|
| 34 |
+
and conversions to other media types.
|
| 35 |
+
|
| 36 |
+
"Work" shall mean the work of authorship, whether in Source or
|
| 37 |
+
Object form, made available under the License, as indicated by a
|
| 38 |
+
copyright notice that is included in or attached to the work
|
| 39 |
+
(an example is provided in the Appendix below).
|
| 40 |
+
|
| 41 |
+
"Derivative Works" shall mean any work, whether in Source or Object
|
| 42 |
+
form, that is based on (or derived from) the Work and for which the
|
| 43 |
+
editorial revisions, annotations, elaborations, or other modifications
|
| 44 |
+
represent, as a whole, an original work of authorship. For the purposes
|
| 45 |
+
of this License, Derivative Works shall not include works that remain
|
| 46 |
+
separable from, or merely link (or bind by name) to the interfaces of,
|
| 47 |
+
the Work and Derivative Works thereof.
|
| 48 |
+
|
| 49 |
+
"Contribution" shall mean any work of authorship, including
|
| 50 |
+
the original version of the Work and any modifications or additions
|
| 51 |
+
to that Work or Derivative Works thereof, that is intentionally
|
| 52 |
+
submitted to Licensor for inclusion in the Work by the copyright owner
|
| 53 |
+
or by an individual or Legal Entity authorized to submit on behalf of
|
| 54 |
+
the copyright owner. For the purposes of this definition, "submitted"
|
| 55 |
+
means any form of electronic, verbal, or written communication sent
|
| 56 |
+
to the Licensor or its representatives, including but not limited to
|
| 57 |
+
communication on electronic mailing lists, source code control systems,
|
| 58 |
+
and issue tracking systems that are managed by, or on behalf of, the
|
| 59 |
+
Licensor for the purpose of discussing and improving the Work, but
|
| 60 |
+
excluding communication that is conspicuously marked or otherwise
|
| 61 |
+
designated in writing by the copyright owner as "Not a Contribution."
|
| 62 |
+
|
| 63 |
+
"Contributor" shall mean Licensor and any individual or Legal Entity
|
| 64 |
+
on behalf of whom a Contribution has been received by Licensor and
|
| 65 |
+
subsequently incorporated within the Work.
|
| 66 |
+
|
| 67 |
+
2. Grant of Copyright License. Subject to the terms and conditions of
|
| 68 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 69 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 70 |
+
copyright license to reproduce, prepare Derivative Works of,
|
| 71 |
+
publicly display, publicly perform, sublicense, and distribute the
|
| 72 |
+
Work and such Derivative Works in Source or Object form.
|
| 73 |
+
|
| 74 |
+
3. Grant of Patent License. Subject to the terms and conditions of
|
| 75 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 76 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 77 |
+
(except as stated in this section) patent license to make, have made,
|
| 78 |
+
use, offer to sell, sell, import, and otherwise transfer the Work,
|
| 79 |
+
where such license applies only to those patent claims licensable
|
| 80 |
+
by such Contributor that are necessarily infringed by their
|
| 81 |
+
Contribution(s) alone or by combination of their Contribution(s)
|
| 82 |
+
with the Work to which such Contribution(s) was submitted. If You
|
| 83 |
+
institute patent litigation against any entity (including a
|
| 84 |
+
cross-claim or counterclaim in a lawsuit) alleging that the Work
|
| 85 |
+
or a Contribution incorporated within the Work constitutes direct
|
| 86 |
+
or contributory patent infringement, then any patent licenses
|
| 87 |
+
granted to You under this License for that Work shall terminate
|
| 88 |
+
as of the date such litigation is filed.
|
| 89 |
+
|
| 90 |
+
4. Redistribution. You may reproduce and distribute copies of the
|
| 91 |
+
Work or Derivative Works thereof in any medium, with or without
|
| 92 |
+
modifications, and in Source or Object form, provided that You
|
| 93 |
+
meet the following conditions:
|
| 94 |
+
|
| 95 |
+
(a) You must give any other recipients of the Work or
|
| 96 |
+
Derivative Works a copy of this License; and
|
| 97 |
+
|
| 98 |
+
(b) You must cause any modified files to carry prominent notices
|
| 99 |
+
stating that You changed the files; and
|
| 100 |
+
|
| 101 |
+
(c) You must retain, in the Source form of any Derivative Works
|
| 102 |
+
that You distribute, all copyright, patent, trademark, and
|
| 103 |
+
attribution notices from the Source form of the Work,
|
| 104 |
+
excluding those notices that do not pertain to any part of
|
| 105 |
+
the Derivative Works; and
|
| 106 |
+
|
| 107 |
+
(d) If the Work includes a "NOTICE" text file as part of its
|
| 108 |
+
distribution, then any Derivative Works that You distribute must
|
| 109 |
+
include a readable copy of the attribution notices contained
|
| 110 |
+
within such NOTICE file, excluding those notices that do not
|
| 111 |
+
pertain to any part of the Derivative Works, in at least one
|
| 112 |
+
of the following places: within a NOTICE text file distributed
|
| 113 |
+
as part of the Derivative Works; within the Source form or
|
| 114 |
+
documentation, if provided along with the Derivative Works; or,
|
| 115 |
+
within a display generated by the Derivative Works, if and
|
| 116 |
+
wherever such third-party notices normally appear. The contents
|
| 117 |
+
of the NOTICE file are for informational purposes only and
|
| 118 |
+
do not modify the License. You may add Your own attribution
|
| 119 |
+
notices within Derivative Works that You distribute, alongside
|
| 120 |
+
or as an addendum to the NOTICE text from the Work, provided
|
| 121 |
+
that such additional attribution notices cannot be construed
|
| 122 |
+
as modifying the License.
|
| 123 |
+
|
| 124 |
+
You may add Your own copyright statement to Your modifications and
|
| 125 |
+
may provide additional or different license terms and conditions
|
| 126 |
+
for use, reproduction, or distribution of Your modifications, or
|
| 127 |
+
for any such Derivative Works as a whole, provided Your use,
|
| 128 |
+
reproduction, and distribution of the Work otherwise complies with
|
| 129 |
+
the conditions stated in this License.
|
| 130 |
+
|
| 131 |
+
5. Submission of Contributions. Unless You explicitly state otherwise,
|
| 132 |
+
any Contribution intentionally submitted for inclusion in the Work
|
| 133 |
+
by You to the Licensor shall be under the terms and conditions of
|
| 134 |
+
this License, without any additional terms or conditions.
|
| 135 |
+
Notwithstanding the above, nothing herein shall supersede or modify
|
| 136 |
+
the terms of any separate license agreement you may have executed
|
| 137 |
+
with Licensor regarding such Contributions.
|
| 138 |
+
|
| 139 |
+
6. Trademarks. This License does not grant permission to use the trade
|
| 140 |
+
names, trademarks, service marks, or product names of the Licensor,
|
| 141 |
+
except as required for reasonable and customary use in describing the
|
| 142 |
+
origin of the Work and reproducing the content of the NOTICE file.
|
| 143 |
+
|
| 144 |
+
7. Disclaimer of Warranty. Unless required by applicable law or
|
| 145 |
+
agreed to in writing, Licensor provides the Work (and each
|
| 146 |
+
Contributor provides its Contributions) on an "AS IS" BASIS,
|
| 147 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
| 148 |
+
implied, including, without limitation, any warranties or conditions
|
| 149 |
+
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
|
| 150 |
+
PARTICULAR PURPOSE. You are solely responsible for determining the
|
| 151 |
+
appropriateness of using or redistributing the Work and assume any
|
| 152 |
+
risks associated with Your exercise of permissions under this License.
|
| 153 |
+
|
| 154 |
+
8. Limitation of Liability. In no event and under no legal theory,
|
| 155 |
+
whether in tort (including negligence), contract, or otherwise,
|
| 156 |
+
unless required by applicable law (such as deliberate and grossly
|
| 157 |
+
negligent acts) or agreed to in writing, shall any Contributor be
|
| 158 |
+
liable to You for damages, including any direct, indirect, special,
|
| 159 |
+
incidental, or consequential damages of any character arising as a
|
| 160 |
+
result of this License or out of the use or inability to use the
|
| 161 |
+
Work (including but not limited to damages for loss of goodwill,
|
| 162 |
+
work stoppage, computer failure or malfunction, or any and all
|
| 163 |
+
other commercial damages or losses), even if such Contributor
|
| 164 |
+
has been advised of the possibility of such damages.
|
| 165 |
+
|
| 166 |
+
9. Accepting Warranty or Additional Liability. While redistributing
|
| 167 |
+
the Work or Derivative Works thereof, You may choose to offer,
|
| 168 |
+
and charge a fee for, acceptance of support, warranty, indemnity,
|
| 169 |
+
or other liability obligations and/or rights consistent with this
|
| 170 |
+
License. However, in accepting such obligations, You may act only
|
| 171 |
+
on Your own behalf and on Your sole responsibility, not on behalf
|
| 172 |
+
of any other Contributor, and only if You agree to indemnify,
|
| 173 |
+
defend, and hold each Contributor harmless for any liability
|
| 174 |
+
incurred by, or claims asserted against, such Contributor by reason
|
| 175 |
+
of your accepting any such warranty or additional liability.
|
| 176 |
+
|
| 177 |
+
END OF TERMS AND CONDITIONS
|
| 178 |
+
|
| 179 |
+
APPENDIX: How to apply the Apache License to your work.
|
| 180 |
+
|
| 181 |
+
To apply the Apache License to your work, attach the following
|
| 182 |
+
boilerplate notice, with the fields enclosed by brackets "[]"
|
| 183 |
+
replaced with your own identifying information. (Don't include
|
| 184 |
+
the brackets!) The text should be enclosed in the appropriate
|
| 185 |
+
comment syntax for the file format. We also recommend that a
|
| 186 |
+
file or class name and description of purpose be included on the
|
| 187 |
+
same "printed page" as the copyright notice for easier
|
| 188 |
+
identification within third-party archives.
|
| 189 |
+
|
| 190 |
+
Copyright [yyyy] [name of copyright owner]
|
| 191 |
+
|
| 192 |
+
Licensed under the Apache License, Version 2.0 (the "License");
|
| 193 |
+
you may not use this file except in compliance with the License.
|
| 194 |
+
You may obtain a copy of the License at
|
| 195 |
+
|
| 196 |
+
http://www.apache.org/licenses/LICENSE-2.0
|
| 197 |
+
|
| 198 |
+
Unless required by applicable law or agreed to in writing, software
|
| 199 |
+
distributed under the License is distributed on an "AS IS" BASIS,
|
| 200 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
| 201 |
+
See the License for the specific language governing permissions and
|
| 202 |
+
limitations under the License.
|
NOTICE
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
SenseNova-Loop-9B-Preview MLX builds by NetProjects (Internet Projects Ltd)
|
| 2 |
+
|
| 3 |
+
This repository contains a modified version of SenseNova-Loop-9B-Preview by SenseNova
|
| 4 |
+
(https://huggingface.co/sensenova/SenseNova-Loop-9B-Preview, revision 9937e91c7301b7dcbb5bf17e3e4e95c499a11e99),
|
| 5 |
+
licensed under the Apache License, Version 2.0 (see LICENSE).
|
| 6 |
+
|
| 7 |
+
Modifications (7 November 2026): the weights were converted to Apple MLX format and quantized (4-bit);
|
| 8 |
+
the vision encoder and the MTP head are not included. sensenova_loop.py is a new MLX implementation of the
|
| 9 |
+
SenseNova-Loop architecture written by NetProjects, also under Apache-2.0.
|
| 10 |
+
|
| 11 |
+
Not affiliated with SenseTime / SenseNova.
|
README.md
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: sensenova/SenseNova-Loop-9B-Preview
|
| 4 |
+
base_model_relation: quantized
|
| 5 |
+
library_name: mlx
|
| 6 |
+
pipeline_tag: text-generation
|
| 7 |
+
tags: [mlx, sensenova, sensenova-loop, looped-transformer, mhc, qwen3_5, apple-silicon]
|
| 8 |
+
---
|
| 9 |
+
# SenseNova-Loop-9B-Preview — MLX 4-bit
|
| 10 |
+
|
| 11 |
+
A 4-bit MLX conversion (affine, group size 64) of [sensenova/SenseNova-Loop-9B-Preview](https://huggingface.co/sensenova/SenseNova-Loop-9B-Preview) (revision `9937e91`, released 30 October 2026) for Apple Silicon. 4.7 GB of weights. Text model; image input is not part of this build yet. The smallest build. It follows the reference less closely than 8-bit and 6-bit (89% same top token); pick 6-bit or 8-bit when quality matters most.
|
| 12 |
+
|
| 13 |
+
Published by [NetProjects](https://netprojects.com) (Internet Projects Ltd). Not affiliated with SenseTime / SenseNova.
|
| 14 |
+
|
| 15 |
+
## A faithful port of the looped architecture
|
| 16 |
+
|
| 17 |
+
SenseNova-Loop is a Qwen3.5-based model with two additions, and this build runs both exactly as SenseNova's own code does:
|
| 18 |
+
|
| 19 |
+
- **Looped execution:** layers 0–3 run once, layers 4–27 run **twice** with shared weights, then layers 28–31: 56 layer passes per token. Every pass has its own cache slot (KV cache for full attention, conv and recurrent state for Gated DeltaNet).
|
| 20 |
+
- **Two-stream mHC** on layers 4–27: the hidden state is split into two residual streams that are routed by small float32 routers (pre / post gates and a 2×2 Sinkhorn mixing matrix, 20 iterations). The 96 router tensors stay in float32 and are never quantized.
|
| 21 |
+
|
| 22 |
+
The model code is `sensenova_loop.py`, included in this repo and loaded by mlx-lm through `trust_remote_code=True`.
|
| 23 |
+
|
| 24 |
+
## Results (MacBook Pro M5 Max, 128 GB, 7 November 2026)
|
| 25 |
+
|
| 26 |
+
Agreement with SenseNova's own PyTorch code (Transformers 5.18, bf16) on 16 prompts, every position:
|
| 27 |
+
|
| 28 |
+
| | 8-bit | 6-bit | 4-bit |
|
| 29 |
+
|---|---|---|---|
|
| 30 |
+
| Same top token as the reference | 97.6% | 97.2% | 89.2% |
|
| 31 |
+
| KL divergence at the last token | 0.0011 | 0.0037 | 0.064 |
|
| 32 |
+
| Size | 8.9 GB | 6.8 GB | 4.7 GB |
|
| 33 |
+
|
| 34 |
+
Short public benchmark (same harness for every build; first 300 ARC-Challenge test questions, log-likelihood over the 4 choices; first 50 GSM8K test questions, chat template, thinking off, greedy):
|
| 35 |
+
|
| 36 |
+
| | bf16 | 8-bit | 6-bit | 4-bit |
|
| 37 |
+
|---|---|---|---|---|
|
| 38 |
+
| ARC-Challenge acc / acc_norm | 48.7 / 51.0 | 48.7 / 50.3 | 47.7 / 49.0 | 45.7 / 45.0 |
|
| 39 |
+
| GSM8K (50) | – | 94.0 | 88.0 | 92.0 |
|
| 40 |
+
|
| 41 |
+
GSM8K on 50 questions moves 2 points per question.
|
| 42 |
+
|
| 43 |
+
Decode speed for this build: about 22 tokens/s on the M5 Max (short run). The loop doubles the work of the middle 24 layers, so it is slower than a plain 9B model; that is the cost of the extra depth.
|
| 44 |
+
|
| 45 |
+
## Use
|
| 46 |
+
|
| 47 |
+
```bash
|
| 48 |
+
pip install -U mlx-lm
|
| 49 |
+
```
|
| 50 |
+
|
| 51 |
+
```python
|
| 52 |
+
from mlx_lm import load, generate
|
| 53 |
+
|
| 54 |
+
model, tokenizer = load("netprojects/SenseNova-Loop-9B-Preview-MLX-4bit", trust_remote_code=True)
|
| 55 |
+
messages = [{"role": "user", "content": "Explain in two sentences why the sky is blue."}]
|
| 56 |
+
prompt = tokenizer.apply_chat_template(messages, add_generation_prompt=True, enable_thinking=False)
|
| 57 |
+
print(generate(model, tokenizer, prompt=prompt, max_tokens=256))
|
| 58 |
+
```
|
| 59 |
+
|
| 60 |
+
Set `enable_thinking=True` for thinking mode. Server: `mlx_lm.server --model netprojects/SenseNova-Loop-9B-Preview-MLX-4bit --trust-remote-code`.
|
| 61 |
+
|
| 62 |
+
**Other sizes:** [8-bit](https://huggingface.co/netprojects/SenseNova-Loop-9B-Preview-MLX-8bit) (8.9 GB) · [6-bit](https://huggingface.co/netprojects/SenseNova-Loop-9B-Preview-MLX-6bit) (6.8 GB)
|
| 63 |
+
|
| 64 |
+
## License
|
| 65 |
+
|
| 66 |
+
SenseNova-Loop-9B-Preview is released by SenseNova under the **Apache License 2.0** (`LICENSE` included). This repository contains converted and quantized weights of that model (modified from the original: converted to MLX format and quantized to 4-bit) and an MLX implementation of its architecture, distributed under the same license.
|
| 67 |
+
|
| 68 |
+
Credits: SenseNova / SenseTime (SenseNova-Loop-9B-Preview), Qwen team (Qwen3.5), Apple MLX and mlx-lm. Conversion and MLX port by Paris Michalas, Partner at [NetProjects](https://netprojects.com). Not affiliated with SenseTime / SenseNova.
|
| 69 |
+
|
| 70 |
+
**More NetProjects MLX builds:** [NetProjects MLX builds collection](https://huggingface.co/collections/netprojects/netprojects-mlx-builds-6ae3b2683a59a1b4fc4c0c3a) · [all models at huggingface.co/netprojects](https://huggingface.co/netprojects)
|
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 +1,134 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"SenseNovaLoopForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"attn_implementation": "sdpa",
|
| 6 |
+
"bos_token_id": null,
|
| 7 |
+
"dtype": "bfloat16",
|
| 8 |
+
"eos_token_id": [
|
| 9 |
+
248044,
|
| 10 |
+
248046
|
| 11 |
+
],
|
| 12 |
+
"hidden_size": 4096,
|
| 13 |
+
"image_token_id": 248056,
|
| 14 |
+
"loop_settings": {
|
| 15 |
+
"counts": [
|
| 16 |
+
2
|
| 17 |
+
],
|
| 18 |
+
"end": 28,
|
| 19 |
+
"eval_count": 2,
|
| 20 |
+
"start": 4
|
| 21 |
+
},
|
| 22 |
+
"mhc": {
|
| 23 |
+
"end": 28,
|
| 24 |
+
"fixed_identity": false,
|
| 25 |
+
"iterations": 20,
|
| 26 |
+
"pre_scale": 1.0,
|
| 27 |
+
"start": 4,
|
| 28 |
+
"streams": 2
|
| 29 |
+
},
|
| 30 |
+
"model_name": "SenseNova-Loop-9B-Preview",
|
| 31 |
+
"model_type": "sensenova_loop",
|
| 32 |
+
"pad_token_id": 248044,
|
| 33 |
+
"patch_conv3d": true,
|
| 34 |
+
"quantization": {
|
| 35 |
+
"group_size": 64,
|
| 36 |
+
"bits": 4,
|
| 37 |
+
"mode": "affine"
|
| 38 |
+
},
|
| 39 |
+
"quantization_config": {
|
| 40 |
+
"group_size": 64,
|
| 41 |
+
"bits": 4,
|
| 42 |
+
"mode": "affine"
|
| 43 |
+
},
|
| 44 |
+
"text_config": {
|
| 45 |
+
"attention_bias": false,
|
| 46 |
+
"attention_dropout": 0.0,
|
| 47 |
+
"attn_output_gate": true,
|
| 48 |
+
"bos_token_id": null,
|
| 49 |
+
"dtype": "bfloat16",
|
| 50 |
+
"eos_token_id": 248044,
|
| 51 |
+
"full_attention_interval": 4,
|
| 52 |
+
"head_dim": 256,
|
| 53 |
+
"hidden_act": "silu",
|
| 54 |
+
"hidden_size": 4096,
|
| 55 |
+
"initializer_range": 0.02,
|
| 56 |
+
"intermediate_size": 12288,
|
| 57 |
+
"layer_types": [
|
| 58 |
+
"linear_attention",
|
| 59 |
+
"linear_attention",
|
| 60 |
+
"linear_attention",
|
| 61 |
+
"full_attention",
|
| 62 |
+
"linear_attention",
|
| 63 |
+
"linear_attention",
|
| 64 |
+
"linear_attention",
|
| 65 |
+
"full_attention",
|
| 66 |
+
"linear_attention",
|
| 67 |
+
"linear_attention",
|
| 68 |
+
"linear_attention",
|
| 69 |
+
"full_attention",
|
| 70 |
+
"linear_attention",
|
| 71 |
+
"linear_attention",
|
| 72 |
+
"linear_attention",
|
| 73 |
+
"full_attention",
|
| 74 |
+
"linear_attention",
|
| 75 |
+
"linear_attention",
|
| 76 |
+
"linear_attention",
|
| 77 |
+
"full_attention",
|
| 78 |
+
"linear_attention",
|
| 79 |
+
"linear_attention",
|
| 80 |
+
"linear_attention",
|
| 81 |
+
"full_attention",
|
| 82 |
+
"linear_attention",
|
| 83 |
+
"linear_attention",
|
| 84 |
+
"linear_attention",
|
| 85 |
+
"full_attention",
|
| 86 |
+
"linear_attention",
|
| 87 |
+
"linear_attention",
|
| 88 |
+
"linear_attention",
|
| 89 |
+
"full_attention"
|
| 90 |
+
],
|
| 91 |
+
"linear_conv_kernel_dim": 4,
|
| 92 |
+
"linear_key_head_dim": 128,
|
| 93 |
+
"linear_num_key_heads": 16,
|
| 94 |
+
"linear_num_value_heads": 32,
|
| 95 |
+
"linear_value_head_dim": 128,
|
| 96 |
+
"mamba_ssm_dtype": "float32",
|
| 97 |
+
"max_position_embeddings": 262144,
|
| 98 |
+
"mlp_only_layers": [],
|
| 99 |
+
"model_type": "qwen3_5_text",
|
| 100 |
+
"mtp_num_hidden_layers": 1,
|
| 101 |
+
"mtp_use_dedicated_embeddings": false,
|
| 102 |
+
"num_attention_heads": 16,
|
| 103 |
+
"num_hidden_layers": 32,
|
| 104 |
+
"num_key_value_heads": 4,
|
| 105 |
+
"pad_token_id": 248044,
|
| 106 |
+
"partial_rotary_factor": 0.25,
|
| 107 |
+
"rms_norm_eps": 1e-06,
|
| 108 |
+
"rope_parameters": {
|
| 109 |
+
"mrope_interleaved": true,
|
| 110 |
+
"mrope_section": [
|
| 111 |
+
11,
|
| 112 |
+
11,
|
| 113 |
+
10
|
| 114 |
+
],
|
| 115 |
+
"partial_rotary_factor": 0.25,
|
| 116 |
+
"rope_theta": 10000000,
|
| 117 |
+
"type": "default"
|
| 118 |
+
},
|
| 119 |
+
"tie_word_embeddings": false,
|
| 120 |
+
"use_cache": true,
|
| 121 |
+
"vocab_size": 248320
|
| 122 |
+
},
|
| 123 |
+
"tie_word_embeddings": false,
|
| 124 |
+
"use_cache": true,
|
| 125 |
+
"video_token_id": 248057,
|
| 126 |
+
"vision_end_token_id": 248054,
|
| 127 |
+
"vision_start_token_id": 248053,
|
| 128 |
+
"model_file": "sensenova_loop.py",
|
| 129 |
+
"sensenova_mlx": {
|
| 130 |
+
"source": "sensenova/SenseNova-Loop-9B-Preview",
|
| 131 |
+
"revision": "9937e91c7301b7dcbb5bf17e3e4e95c499a11e99",
|
| 132 |
+
"text_only": true
|
| 133 |
+
}
|
| 134 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": [
|
| 4 |
+
248044,
|
| 5 |
+
248046
|
| 6 |
+
],
|
| 7 |
+
"transformers_version": "5.18.0",
|
| 8 |
+
"use_cache": true,
|
| 9 |
+
"temperature": 1.0,
|
| 10 |
+
"do_sample": true,
|
| 11 |
+
"top_p": 0.95
|
| 12 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:608aee419eb56d54863fd4388e552110f3c588f9f040375adef67805ccf738fb
|
| 3 |
+
size 5050758107
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,1031 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_size": 5050626048,
|
| 4 |
+
"total_parameters": 8956947840
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"language_model.lm_head.biases": "model.safetensors",
|
| 8 |
+
"language_model.lm_head.scales": "model.safetensors",
|
| 9 |
+
"language_model.lm_head.weight": "model.safetensors",
|
| 10 |
+
"language_model.model.embed_tokens.biases": "model.safetensors",
|
| 11 |
+
"language_model.model.embed_tokens.scales": "model.safetensors",
|
| 12 |
+
"language_model.model.embed_tokens.weight": "model.safetensors",
|
| 13 |
+
"language_model.model.layers.0.input_layernorm.weight": "model.safetensors",
|
| 14 |
+
"language_model.model.layers.0.linear_attn.A_log": "model.safetensors",
|
| 15 |
+
"language_model.model.layers.0.linear_attn.conv1d.weight": "model.safetensors",
|
| 16 |
+
"language_model.model.layers.0.linear_attn.dt_bias": "model.safetensors",
|
| 17 |
+
"language_model.model.layers.0.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 18 |
+
"language_model.model.layers.0.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 19 |
+
"language_model.model.layers.0.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 20 |
+
"language_model.model.layers.0.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 21 |
+
"language_model.model.layers.0.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 22 |
+
"language_model.model.layers.0.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 23 |
+
"language_model.model.layers.0.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 24 |
+
"language_model.model.layers.0.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 25 |
+
"language_model.model.layers.0.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 26 |
+
"language_model.model.layers.0.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 27 |
+
"language_model.model.layers.0.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 28 |
+
"language_model.model.layers.0.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 29 |
+
"language_model.model.layers.0.linear_attn.norm.weight": "model.safetensors",
|
| 30 |
+
"language_model.model.layers.0.linear_attn.out_proj.biases": "model.safetensors",
|
| 31 |
+
"language_model.model.layers.0.linear_attn.out_proj.scales": "model.safetensors",
|
| 32 |
+
"language_model.model.layers.0.linear_attn.out_proj.weight": "model.safetensors",
|
| 33 |
+
"language_model.model.layers.0.mlp.down_proj.biases": "model.safetensors",
|
| 34 |
+
"language_model.model.layers.0.mlp.down_proj.scales": "model.safetensors",
|
| 35 |
+
"language_model.model.layers.0.mlp.down_proj.weight": "model.safetensors",
|
| 36 |
+
"language_model.model.layers.0.mlp.gate_proj.biases": "model.safetensors",
|
| 37 |
+
"language_model.model.layers.0.mlp.gate_proj.scales": "model.safetensors",
|
| 38 |
+
"language_model.model.layers.0.mlp.gate_proj.weight": "model.safetensors",
|
| 39 |
+
"language_model.model.layers.0.mlp.up_proj.biases": "model.safetensors",
|
| 40 |
+
"language_model.model.layers.0.mlp.up_proj.scales": "model.safetensors",
|
| 41 |
+
"language_model.model.layers.0.mlp.up_proj.weight": "model.safetensors",
|
| 42 |
+
"language_model.model.layers.0.post_attention_layernorm.weight": "model.safetensors",
|
| 43 |
+
"language_model.model.layers.1.input_layernorm.weight": "model.safetensors",
|
| 44 |
+
"language_model.model.layers.1.linear_attn.A_log": "model.safetensors",
|
| 45 |
+
"language_model.model.layers.1.linear_attn.conv1d.weight": "model.safetensors",
|
| 46 |
+
"language_model.model.layers.1.linear_attn.dt_bias": "model.safetensors",
|
| 47 |
+
"language_model.model.layers.1.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 48 |
+
"language_model.model.layers.1.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 49 |
+
"language_model.model.layers.1.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 50 |
+
"language_model.model.layers.1.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 51 |
+
"language_model.model.layers.1.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 52 |
+
"language_model.model.layers.1.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 53 |
+
"language_model.model.layers.1.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 54 |
+
"language_model.model.layers.1.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 55 |
+
"language_model.model.layers.1.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 56 |
+
"language_model.model.layers.1.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 57 |
+
"language_model.model.layers.1.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 58 |
+
"language_model.model.layers.1.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 59 |
+
"language_model.model.layers.1.linear_attn.norm.weight": "model.safetensors",
|
| 60 |
+
"language_model.model.layers.1.linear_attn.out_proj.biases": "model.safetensors",
|
| 61 |
+
"language_model.model.layers.1.linear_attn.out_proj.scales": "model.safetensors",
|
| 62 |
+
"language_model.model.layers.1.linear_attn.out_proj.weight": "model.safetensors",
|
| 63 |
+
"language_model.model.layers.1.mlp.down_proj.biases": "model.safetensors",
|
| 64 |
+
"language_model.model.layers.1.mlp.down_proj.scales": "model.safetensors",
|
| 65 |
+
"language_model.model.layers.1.mlp.down_proj.weight": "model.safetensors",
|
| 66 |
+
"language_model.model.layers.1.mlp.gate_proj.biases": "model.safetensors",
|
| 67 |
+
"language_model.model.layers.1.mlp.gate_proj.scales": "model.safetensors",
|
| 68 |
+
"language_model.model.layers.1.mlp.gate_proj.weight": "model.safetensors",
|
| 69 |
+
"language_model.model.layers.1.mlp.up_proj.biases": "model.safetensors",
|
| 70 |
+
"language_model.model.layers.1.mlp.up_proj.scales": "model.safetensors",
|
| 71 |
+
"language_model.model.layers.1.mlp.up_proj.weight": "model.safetensors",
|
| 72 |
+
"language_model.model.layers.1.post_attention_layernorm.weight": "model.safetensors",
|
| 73 |
+
"language_model.model.layers.10.input_layernorm.weight": "model.safetensors",
|
| 74 |
+
"language_model.model.layers.10.linear_attn.A_log": "model.safetensors",
|
| 75 |
+
"language_model.model.layers.10.linear_attn.conv1d.weight": "model.safetensors",
|
| 76 |
+
"language_model.model.layers.10.linear_attn.dt_bias": "model.safetensors",
|
| 77 |
+
"language_model.model.layers.10.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 78 |
+
"language_model.model.layers.10.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 79 |
+
"language_model.model.layers.10.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 80 |
+
"language_model.model.layers.10.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 81 |
+
"language_model.model.layers.10.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 82 |
+
"language_model.model.layers.10.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 83 |
+
"language_model.model.layers.10.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 84 |
+
"language_model.model.layers.10.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 85 |
+
"language_model.model.layers.10.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 86 |
+
"language_model.model.layers.10.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 87 |
+
"language_model.model.layers.10.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 88 |
+
"language_model.model.layers.10.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 89 |
+
"language_model.model.layers.10.linear_attn.norm.weight": "model.safetensors",
|
| 90 |
+
"language_model.model.layers.10.linear_attn.out_proj.biases": "model.safetensors",
|
| 91 |
+
"language_model.model.layers.10.linear_attn.out_proj.scales": "model.safetensors",
|
| 92 |
+
"language_model.model.layers.10.linear_attn.out_proj.weight": "model.safetensors",
|
| 93 |
+
"language_model.model.layers.10.mhc_attn.bias": "model.safetensors",
|
| 94 |
+
"language_model.model.layers.10.mhc_attn.weight": "model.safetensors",
|
| 95 |
+
"language_model.model.layers.10.mhc_ffn.bias": "model.safetensors",
|
| 96 |
+
"language_model.model.layers.10.mhc_ffn.weight": "model.safetensors",
|
| 97 |
+
"language_model.model.layers.10.mlp.down_proj.biases": "model.safetensors",
|
| 98 |
+
"language_model.model.layers.10.mlp.down_proj.scales": "model.safetensors",
|
| 99 |
+
"language_model.model.layers.10.mlp.down_proj.weight": "model.safetensors",
|
| 100 |
+
"language_model.model.layers.10.mlp.gate_proj.biases": "model.safetensors",
|
| 101 |
+
"language_model.model.layers.10.mlp.gate_proj.scales": "model.safetensors",
|
| 102 |
+
"language_model.model.layers.10.mlp.gate_proj.weight": "model.safetensors",
|
| 103 |
+
"language_model.model.layers.10.mlp.up_proj.biases": "model.safetensors",
|
| 104 |
+
"language_model.model.layers.10.mlp.up_proj.scales": "model.safetensors",
|
| 105 |
+
"language_model.model.layers.10.mlp.up_proj.weight": "model.safetensors",
|
| 106 |
+
"language_model.model.layers.10.post_attention_layernorm.weight": "model.safetensors",
|
| 107 |
+
"language_model.model.layers.11.input_layernorm.weight": "model.safetensors",
|
| 108 |
+
"language_model.model.layers.11.mhc_attn.bias": "model.safetensors",
|
| 109 |
+
"language_model.model.layers.11.mhc_attn.weight": "model.safetensors",
|
| 110 |
+
"language_model.model.layers.11.mhc_ffn.bias": "model.safetensors",
|
| 111 |
+
"language_model.model.layers.11.mhc_ffn.weight": "model.safetensors",
|
| 112 |
+
"language_model.model.layers.11.mlp.down_proj.biases": "model.safetensors",
|
| 113 |
+
"language_model.model.layers.11.mlp.down_proj.scales": "model.safetensors",
|
| 114 |
+
"language_model.model.layers.11.mlp.down_proj.weight": "model.safetensors",
|
| 115 |
+
"language_model.model.layers.11.mlp.gate_proj.biases": "model.safetensors",
|
| 116 |
+
"language_model.model.layers.11.mlp.gate_proj.scales": "model.safetensors",
|
| 117 |
+
"language_model.model.layers.11.mlp.gate_proj.weight": "model.safetensors",
|
| 118 |
+
"language_model.model.layers.11.mlp.up_proj.biases": "model.safetensors",
|
| 119 |
+
"language_model.model.layers.11.mlp.up_proj.scales": "model.safetensors",
|
| 120 |
+
"language_model.model.layers.11.mlp.up_proj.weight": "model.safetensors",
|
| 121 |
+
"language_model.model.layers.11.post_attention_layernorm.weight": "model.safetensors",
|
| 122 |
+
"language_model.model.layers.11.self_attn.k_norm.weight": "model.safetensors",
|
| 123 |
+
"language_model.model.layers.11.self_attn.k_proj.biases": "model.safetensors",
|
| 124 |
+
"language_model.model.layers.11.self_attn.k_proj.scales": "model.safetensors",
|
| 125 |
+
"language_model.model.layers.11.self_attn.k_proj.weight": "model.safetensors",
|
| 126 |
+
"language_model.model.layers.11.self_attn.o_proj.biases": "model.safetensors",
|
| 127 |
+
"language_model.model.layers.11.self_attn.o_proj.scales": "model.safetensors",
|
| 128 |
+
"language_model.model.layers.11.self_attn.o_proj.weight": "model.safetensors",
|
| 129 |
+
"language_model.model.layers.11.self_attn.q_norm.weight": "model.safetensors",
|
| 130 |
+
"language_model.model.layers.11.self_attn.q_proj.biases": "model.safetensors",
|
| 131 |
+
"language_model.model.layers.11.self_attn.q_proj.scales": "model.safetensors",
|
| 132 |
+
"language_model.model.layers.11.self_attn.q_proj.weight": "model.safetensors",
|
| 133 |
+
"language_model.model.layers.11.self_attn.v_proj.biases": "model.safetensors",
|
| 134 |
+
"language_model.model.layers.11.self_attn.v_proj.scales": "model.safetensors",
|
| 135 |
+
"language_model.model.layers.11.self_attn.v_proj.weight": "model.safetensors",
|
| 136 |
+
"language_model.model.layers.12.input_layernorm.weight": "model.safetensors",
|
| 137 |
+
"language_model.model.layers.12.linear_attn.A_log": "model.safetensors",
|
| 138 |
+
"language_model.model.layers.12.linear_attn.conv1d.weight": "model.safetensors",
|
| 139 |
+
"language_model.model.layers.12.linear_attn.dt_bias": "model.safetensors",
|
| 140 |
+
"language_model.model.layers.12.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 141 |
+
"language_model.model.layers.12.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 142 |
+
"language_model.model.layers.12.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 143 |
+
"language_model.model.layers.12.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 144 |
+
"language_model.model.layers.12.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 145 |
+
"language_model.model.layers.12.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 146 |
+
"language_model.model.layers.12.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 147 |
+
"language_model.model.layers.12.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 148 |
+
"language_model.model.layers.12.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 149 |
+
"language_model.model.layers.12.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 150 |
+
"language_model.model.layers.12.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 151 |
+
"language_model.model.layers.12.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 152 |
+
"language_model.model.layers.12.linear_attn.norm.weight": "model.safetensors",
|
| 153 |
+
"language_model.model.layers.12.linear_attn.out_proj.biases": "model.safetensors",
|
| 154 |
+
"language_model.model.layers.12.linear_attn.out_proj.scales": "model.safetensors",
|
| 155 |
+
"language_model.model.layers.12.linear_attn.out_proj.weight": "model.safetensors",
|
| 156 |
+
"language_model.model.layers.12.mhc_attn.bias": "model.safetensors",
|
| 157 |
+
"language_model.model.layers.12.mhc_attn.weight": "model.safetensors",
|
| 158 |
+
"language_model.model.layers.12.mhc_ffn.bias": "model.safetensors",
|
| 159 |
+
"language_model.model.layers.12.mhc_ffn.weight": "model.safetensors",
|
| 160 |
+
"language_model.model.layers.12.mlp.down_proj.biases": "model.safetensors",
|
| 161 |
+
"language_model.model.layers.12.mlp.down_proj.scales": "model.safetensors",
|
| 162 |
+
"language_model.model.layers.12.mlp.down_proj.weight": "model.safetensors",
|
| 163 |
+
"language_model.model.layers.12.mlp.gate_proj.biases": "model.safetensors",
|
| 164 |
+
"language_model.model.layers.12.mlp.gate_proj.scales": "model.safetensors",
|
| 165 |
+
"language_model.model.layers.12.mlp.gate_proj.weight": "model.safetensors",
|
| 166 |
+
"language_model.model.layers.12.mlp.up_proj.biases": "model.safetensors",
|
| 167 |
+
"language_model.model.layers.12.mlp.up_proj.scales": "model.safetensors",
|
| 168 |
+
"language_model.model.layers.12.mlp.up_proj.weight": "model.safetensors",
|
| 169 |
+
"language_model.model.layers.12.post_attention_layernorm.weight": "model.safetensors",
|
| 170 |
+
"language_model.model.layers.13.input_layernorm.weight": "model.safetensors",
|
| 171 |
+
"language_model.model.layers.13.linear_attn.A_log": "model.safetensors",
|
| 172 |
+
"language_model.model.layers.13.linear_attn.conv1d.weight": "model.safetensors",
|
| 173 |
+
"language_model.model.layers.13.linear_attn.dt_bias": "model.safetensors",
|
| 174 |
+
"language_model.model.layers.13.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 175 |
+
"language_model.model.layers.13.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 176 |
+
"language_model.model.layers.13.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 177 |
+
"language_model.model.layers.13.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 178 |
+
"language_model.model.layers.13.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 179 |
+
"language_model.model.layers.13.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 180 |
+
"language_model.model.layers.13.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 181 |
+
"language_model.model.layers.13.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 182 |
+
"language_model.model.layers.13.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 183 |
+
"language_model.model.layers.13.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 184 |
+
"language_model.model.layers.13.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 185 |
+
"language_model.model.layers.13.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 186 |
+
"language_model.model.layers.13.linear_attn.norm.weight": "model.safetensors",
|
| 187 |
+
"language_model.model.layers.13.linear_attn.out_proj.biases": "model.safetensors",
|
| 188 |
+
"language_model.model.layers.13.linear_attn.out_proj.scales": "model.safetensors",
|
| 189 |
+
"language_model.model.layers.13.linear_attn.out_proj.weight": "model.safetensors",
|
| 190 |
+
"language_model.model.layers.13.mhc_attn.bias": "model.safetensors",
|
| 191 |
+
"language_model.model.layers.13.mhc_attn.weight": "model.safetensors",
|
| 192 |
+
"language_model.model.layers.13.mhc_ffn.bias": "model.safetensors",
|
| 193 |
+
"language_model.model.layers.13.mhc_ffn.weight": "model.safetensors",
|
| 194 |
+
"language_model.model.layers.13.mlp.down_proj.biases": "model.safetensors",
|
| 195 |
+
"language_model.model.layers.13.mlp.down_proj.scales": "model.safetensors",
|
| 196 |
+
"language_model.model.layers.13.mlp.down_proj.weight": "model.safetensors",
|
| 197 |
+
"language_model.model.layers.13.mlp.gate_proj.biases": "model.safetensors",
|
| 198 |
+
"language_model.model.layers.13.mlp.gate_proj.scales": "model.safetensors",
|
| 199 |
+
"language_model.model.layers.13.mlp.gate_proj.weight": "model.safetensors",
|
| 200 |
+
"language_model.model.layers.13.mlp.up_proj.biases": "model.safetensors",
|
| 201 |
+
"language_model.model.layers.13.mlp.up_proj.scales": "model.safetensors",
|
| 202 |
+
"language_model.model.layers.13.mlp.up_proj.weight": "model.safetensors",
|
| 203 |
+
"language_model.model.layers.13.post_attention_layernorm.weight": "model.safetensors",
|
| 204 |
+
"language_model.model.layers.14.input_layernorm.weight": "model.safetensors",
|
| 205 |
+
"language_model.model.layers.14.linear_attn.A_log": "model.safetensors",
|
| 206 |
+
"language_model.model.layers.14.linear_attn.conv1d.weight": "model.safetensors",
|
| 207 |
+
"language_model.model.layers.14.linear_attn.dt_bias": "model.safetensors",
|
| 208 |
+
"language_model.model.layers.14.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 209 |
+
"language_model.model.layers.14.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 210 |
+
"language_model.model.layers.14.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 211 |
+
"language_model.model.layers.14.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 212 |
+
"language_model.model.layers.14.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 213 |
+
"language_model.model.layers.14.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 214 |
+
"language_model.model.layers.14.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 215 |
+
"language_model.model.layers.14.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 216 |
+
"language_model.model.layers.14.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 217 |
+
"language_model.model.layers.14.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 218 |
+
"language_model.model.layers.14.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 219 |
+
"language_model.model.layers.14.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 220 |
+
"language_model.model.layers.14.linear_attn.norm.weight": "model.safetensors",
|
| 221 |
+
"language_model.model.layers.14.linear_attn.out_proj.biases": "model.safetensors",
|
| 222 |
+
"language_model.model.layers.14.linear_attn.out_proj.scales": "model.safetensors",
|
| 223 |
+
"language_model.model.layers.14.linear_attn.out_proj.weight": "model.safetensors",
|
| 224 |
+
"language_model.model.layers.14.mhc_attn.bias": "model.safetensors",
|
| 225 |
+
"language_model.model.layers.14.mhc_attn.weight": "model.safetensors",
|
| 226 |
+
"language_model.model.layers.14.mhc_ffn.bias": "model.safetensors",
|
| 227 |
+
"language_model.model.layers.14.mhc_ffn.weight": "model.safetensors",
|
| 228 |
+
"language_model.model.layers.14.mlp.down_proj.biases": "model.safetensors",
|
| 229 |
+
"language_model.model.layers.14.mlp.down_proj.scales": "model.safetensors",
|
| 230 |
+
"language_model.model.layers.14.mlp.down_proj.weight": "model.safetensors",
|
| 231 |
+
"language_model.model.layers.14.mlp.gate_proj.biases": "model.safetensors",
|
| 232 |
+
"language_model.model.layers.14.mlp.gate_proj.scales": "model.safetensors",
|
| 233 |
+
"language_model.model.layers.14.mlp.gate_proj.weight": "model.safetensors",
|
| 234 |
+
"language_model.model.layers.14.mlp.up_proj.biases": "model.safetensors",
|
| 235 |
+
"language_model.model.layers.14.mlp.up_proj.scales": "model.safetensors",
|
| 236 |
+
"language_model.model.layers.14.mlp.up_proj.weight": "model.safetensors",
|
| 237 |
+
"language_model.model.layers.14.post_attention_layernorm.weight": "model.safetensors",
|
| 238 |
+
"language_model.model.layers.15.input_layernorm.weight": "model.safetensors",
|
| 239 |
+
"language_model.model.layers.15.mhc_attn.bias": "model.safetensors",
|
| 240 |
+
"language_model.model.layers.15.mhc_attn.weight": "model.safetensors",
|
| 241 |
+
"language_model.model.layers.15.mhc_ffn.bias": "model.safetensors",
|
| 242 |
+
"language_model.model.layers.15.mhc_ffn.weight": "model.safetensors",
|
| 243 |
+
"language_model.model.layers.15.mlp.down_proj.biases": "model.safetensors",
|
| 244 |
+
"language_model.model.layers.15.mlp.down_proj.scales": "model.safetensors",
|
| 245 |
+
"language_model.model.layers.15.mlp.down_proj.weight": "model.safetensors",
|
| 246 |
+
"language_model.model.layers.15.mlp.gate_proj.biases": "model.safetensors",
|
| 247 |
+
"language_model.model.layers.15.mlp.gate_proj.scales": "model.safetensors",
|
| 248 |
+
"language_model.model.layers.15.mlp.gate_proj.weight": "model.safetensors",
|
| 249 |
+
"language_model.model.layers.15.mlp.up_proj.biases": "model.safetensors",
|
| 250 |
+
"language_model.model.layers.15.mlp.up_proj.scales": "model.safetensors",
|
| 251 |
+
"language_model.model.layers.15.mlp.up_proj.weight": "model.safetensors",
|
| 252 |
+
"language_model.model.layers.15.post_attention_layernorm.weight": "model.safetensors",
|
| 253 |
+
"language_model.model.layers.15.self_attn.k_norm.weight": "model.safetensors",
|
| 254 |
+
"language_model.model.layers.15.self_attn.k_proj.biases": "model.safetensors",
|
| 255 |
+
"language_model.model.layers.15.self_attn.k_proj.scales": "model.safetensors",
|
| 256 |
+
"language_model.model.layers.15.self_attn.k_proj.weight": "model.safetensors",
|
| 257 |
+
"language_model.model.layers.15.self_attn.o_proj.biases": "model.safetensors",
|
| 258 |
+
"language_model.model.layers.15.self_attn.o_proj.scales": "model.safetensors",
|
| 259 |
+
"language_model.model.layers.15.self_attn.o_proj.weight": "model.safetensors",
|
| 260 |
+
"language_model.model.layers.15.self_attn.q_norm.weight": "model.safetensors",
|
| 261 |
+
"language_model.model.layers.15.self_attn.q_proj.biases": "model.safetensors",
|
| 262 |
+
"language_model.model.layers.15.self_attn.q_proj.scales": "model.safetensors",
|
| 263 |
+
"language_model.model.layers.15.self_attn.q_proj.weight": "model.safetensors",
|
| 264 |
+
"language_model.model.layers.15.self_attn.v_proj.biases": "model.safetensors",
|
| 265 |
+
"language_model.model.layers.15.self_attn.v_proj.scales": "model.safetensors",
|
| 266 |
+
"language_model.model.layers.15.self_attn.v_proj.weight": "model.safetensors",
|
| 267 |
+
"language_model.model.layers.16.input_layernorm.weight": "model.safetensors",
|
| 268 |
+
"language_model.model.layers.16.linear_attn.A_log": "model.safetensors",
|
| 269 |
+
"language_model.model.layers.16.linear_attn.conv1d.weight": "model.safetensors",
|
| 270 |
+
"language_model.model.layers.16.linear_attn.dt_bias": "model.safetensors",
|
| 271 |
+
"language_model.model.layers.16.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 272 |
+
"language_model.model.layers.16.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 273 |
+
"language_model.model.layers.16.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 274 |
+
"language_model.model.layers.16.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 275 |
+
"language_model.model.layers.16.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 276 |
+
"language_model.model.layers.16.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 277 |
+
"language_model.model.layers.16.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 278 |
+
"language_model.model.layers.16.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 279 |
+
"language_model.model.layers.16.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 280 |
+
"language_model.model.layers.16.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 281 |
+
"language_model.model.layers.16.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 282 |
+
"language_model.model.layers.16.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 283 |
+
"language_model.model.layers.16.linear_attn.norm.weight": "model.safetensors",
|
| 284 |
+
"language_model.model.layers.16.linear_attn.out_proj.biases": "model.safetensors",
|
| 285 |
+
"language_model.model.layers.16.linear_attn.out_proj.scales": "model.safetensors",
|
| 286 |
+
"language_model.model.layers.16.linear_attn.out_proj.weight": "model.safetensors",
|
| 287 |
+
"language_model.model.layers.16.mhc_attn.bias": "model.safetensors",
|
| 288 |
+
"language_model.model.layers.16.mhc_attn.weight": "model.safetensors",
|
| 289 |
+
"language_model.model.layers.16.mhc_ffn.bias": "model.safetensors",
|
| 290 |
+
"language_model.model.layers.16.mhc_ffn.weight": "model.safetensors",
|
| 291 |
+
"language_model.model.layers.16.mlp.down_proj.biases": "model.safetensors",
|
| 292 |
+
"language_model.model.layers.16.mlp.down_proj.scales": "model.safetensors",
|
| 293 |
+
"language_model.model.layers.16.mlp.down_proj.weight": "model.safetensors",
|
| 294 |
+
"language_model.model.layers.16.mlp.gate_proj.biases": "model.safetensors",
|
| 295 |
+
"language_model.model.layers.16.mlp.gate_proj.scales": "model.safetensors",
|
| 296 |
+
"language_model.model.layers.16.mlp.gate_proj.weight": "model.safetensors",
|
| 297 |
+
"language_model.model.layers.16.mlp.up_proj.biases": "model.safetensors",
|
| 298 |
+
"language_model.model.layers.16.mlp.up_proj.scales": "model.safetensors",
|
| 299 |
+
"language_model.model.layers.16.mlp.up_proj.weight": "model.safetensors",
|
| 300 |
+
"language_model.model.layers.16.post_attention_layernorm.weight": "model.safetensors",
|
| 301 |
+
"language_model.model.layers.17.input_layernorm.weight": "model.safetensors",
|
| 302 |
+
"language_model.model.layers.17.linear_attn.A_log": "model.safetensors",
|
| 303 |
+
"language_model.model.layers.17.linear_attn.conv1d.weight": "model.safetensors",
|
| 304 |
+
"language_model.model.layers.17.linear_attn.dt_bias": "model.safetensors",
|
| 305 |
+
"language_model.model.layers.17.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 306 |
+
"language_model.model.layers.17.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 307 |
+
"language_model.model.layers.17.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 308 |
+
"language_model.model.layers.17.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 309 |
+
"language_model.model.layers.17.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 310 |
+
"language_model.model.layers.17.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 311 |
+
"language_model.model.layers.17.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 312 |
+
"language_model.model.layers.17.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 313 |
+
"language_model.model.layers.17.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 314 |
+
"language_model.model.layers.17.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 315 |
+
"language_model.model.layers.17.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 316 |
+
"language_model.model.layers.17.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 317 |
+
"language_model.model.layers.17.linear_attn.norm.weight": "model.safetensors",
|
| 318 |
+
"language_model.model.layers.17.linear_attn.out_proj.biases": "model.safetensors",
|
| 319 |
+
"language_model.model.layers.17.linear_attn.out_proj.scales": "model.safetensors",
|
| 320 |
+
"language_model.model.layers.17.linear_attn.out_proj.weight": "model.safetensors",
|
| 321 |
+
"language_model.model.layers.17.mhc_attn.bias": "model.safetensors",
|
| 322 |
+
"language_model.model.layers.17.mhc_attn.weight": "model.safetensors",
|
| 323 |
+
"language_model.model.layers.17.mhc_ffn.bias": "model.safetensors",
|
| 324 |
+
"language_model.model.layers.17.mhc_ffn.weight": "model.safetensors",
|
| 325 |
+
"language_model.model.layers.17.mlp.down_proj.biases": "model.safetensors",
|
| 326 |
+
"language_model.model.layers.17.mlp.down_proj.scales": "model.safetensors",
|
| 327 |
+
"language_model.model.layers.17.mlp.down_proj.weight": "model.safetensors",
|
| 328 |
+
"language_model.model.layers.17.mlp.gate_proj.biases": "model.safetensors",
|
| 329 |
+
"language_model.model.layers.17.mlp.gate_proj.scales": "model.safetensors",
|
| 330 |
+
"language_model.model.layers.17.mlp.gate_proj.weight": "model.safetensors",
|
| 331 |
+
"language_model.model.layers.17.mlp.up_proj.biases": "model.safetensors",
|
| 332 |
+
"language_model.model.layers.17.mlp.up_proj.scales": "model.safetensors",
|
| 333 |
+
"language_model.model.layers.17.mlp.up_proj.weight": "model.safetensors",
|
| 334 |
+
"language_model.model.layers.17.post_attention_layernorm.weight": "model.safetensors",
|
| 335 |
+
"language_model.model.layers.18.input_layernorm.weight": "model.safetensors",
|
| 336 |
+
"language_model.model.layers.18.linear_attn.A_log": "model.safetensors",
|
| 337 |
+
"language_model.model.layers.18.linear_attn.conv1d.weight": "model.safetensors",
|
| 338 |
+
"language_model.model.layers.18.linear_attn.dt_bias": "model.safetensors",
|
| 339 |
+
"language_model.model.layers.18.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 340 |
+
"language_model.model.layers.18.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 341 |
+
"language_model.model.layers.18.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 342 |
+
"language_model.model.layers.18.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 343 |
+
"language_model.model.layers.18.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 344 |
+
"language_model.model.layers.18.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 345 |
+
"language_model.model.layers.18.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 346 |
+
"language_model.model.layers.18.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 347 |
+
"language_model.model.layers.18.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 348 |
+
"language_model.model.layers.18.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 349 |
+
"language_model.model.layers.18.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 350 |
+
"language_model.model.layers.18.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 351 |
+
"language_model.model.layers.18.linear_attn.norm.weight": "model.safetensors",
|
| 352 |
+
"language_model.model.layers.18.linear_attn.out_proj.biases": "model.safetensors",
|
| 353 |
+
"language_model.model.layers.18.linear_attn.out_proj.scales": "model.safetensors",
|
| 354 |
+
"language_model.model.layers.18.linear_attn.out_proj.weight": "model.safetensors",
|
| 355 |
+
"language_model.model.layers.18.mhc_attn.bias": "model.safetensors",
|
| 356 |
+
"language_model.model.layers.18.mhc_attn.weight": "model.safetensors",
|
| 357 |
+
"language_model.model.layers.18.mhc_ffn.bias": "model.safetensors",
|
| 358 |
+
"language_model.model.layers.18.mhc_ffn.weight": "model.safetensors",
|
| 359 |
+
"language_model.model.layers.18.mlp.down_proj.biases": "model.safetensors",
|
| 360 |
+
"language_model.model.layers.18.mlp.down_proj.scales": "model.safetensors",
|
| 361 |
+
"language_model.model.layers.18.mlp.down_proj.weight": "model.safetensors",
|
| 362 |
+
"language_model.model.layers.18.mlp.gate_proj.biases": "model.safetensors",
|
| 363 |
+
"language_model.model.layers.18.mlp.gate_proj.scales": "model.safetensors",
|
| 364 |
+
"language_model.model.layers.18.mlp.gate_proj.weight": "model.safetensors",
|
| 365 |
+
"language_model.model.layers.18.mlp.up_proj.biases": "model.safetensors",
|
| 366 |
+
"language_model.model.layers.18.mlp.up_proj.scales": "model.safetensors",
|
| 367 |
+
"language_model.model.layers.18.mlp.up_proj.weight": "model.safetensors",
|
| 368 |
+
"language_model.model.layers.18.post_attention_layernorm.weight": "model.safetensors",
|
| 369 |
+
"language_model.model.layers.19.input_layernorm.weight": "model.safetensors",
|
| 370 |
+
"language_model.model.layers.19.mhc_attn.bias": "model.safetensors",
|
| 371 |
+
"language_model.model.layers.19.mhc_attn.weight": "model.safetensors",
|
| 372 |
+
"language_model.model.layers.19.mhc_ffn.bias": "model.safetensors",
|
| 373 |
+
"language_model.model.layers.19.mhc_ffn.weight": "model.safetensors",
|
| 374 |
+
"language_model.model.layers.19.mlp.down_proj.biases": "model.safetensors",
|
| 375 |
+
"language_model.model.layers.19.mlp.down_proj.scales": "model.safetensors",
|
| 376 |
+
"language_model.model.layers.19.mlp.down_proj.weight": "model.safetensors",
|
| 377 |
+
"language_model.model.layers.19.mlp.gate_proj.biases": "model.safetensors",
|
| 378 |
+
"language_model.model.layers.19.mlp.gate_proj.scales": "model.safetensors",
|
| 379 |
+
"language_model.model.layers.19.mlp.gate_proj.weight": "model.safetensors",
|
| 380 |
+
"language_model.model.layers.19.mlp.up_proj.biases": "model.safetensors",
|
| 381 |
+
"language_model.model.layers.19.mlp.up_proj.scales": "model.safetensors",
|
| 382 |
+
"language_model.model.layers.19.mlp.up_proj.weight": "model.safetensors",
|
| 383 |
+
"language_model.model.layers.19.post_attention_layernorm.weight": "model.safetensors",
|
| 384 |
+
"language_model.model.layers.19.self_attn.k_norm.weight": "model.safetensors",
|
| 385 |
+
"language_model.model.layers.19.self_attn.k_proj.biases": "model.safetensors",
|
| 386 |
+
"language_model.model.layers.19.self_attn.k_proj.scales": "model.safetensors",
|
| 387 |
+
"language_model.model.layers.19.self_attn.k_proj.weight": "model.safetensors",
|
| 388 |
+
"language_model.model.layers.19.self_attn.o_proj.biases": "model.safetensors",
|
| 389 |
+
"language_model.model.layers.19.self_attn.o_proj.scales": "model.safetensors",
|
| 390 |
+
"language_model.model.layers.19.self_attn.o_proj.weight": "model.safetensors",
|
| 391 |
+
"language_model.model.layers.19.self_attn.q_norm.weight": "model.safetensors",
|
| 392 |
+
"language_model.model.layers.19.self_attn.q_proj.biases": "model.safetensors",
|
| 393 |
+
"language_model.model.layers.19.self_attn.q_proj.scales": "model.safetensors",
|
| 394 |
+
"language_model.model.layers.19.self_attn.q_proj.weight": "model.safetensors",
|
| 395 |
+
"language_model.model.layers.19.self_attn.v_proj.biases": "model.safetensors",
|
| 396 |
+
"language_model.model.layers.19.self_attn.v_proj.scales": "model.safetensors",
|
| 397 |
+
"language_model.model.layers.19.self_attn.v_proj.weight": "model.safetensors",
|
| 398 |
+
"language_model.model.layers.2.input_layernorm.weight": "model.safetensors",
|
| 399 |
+
"language_model.model.layers.2.linear_attn.A_log": "model.safetensors",
|
| 400 |
+
"language_model.model.layers.2.linear_attn.conv1d.weight": "model.safetensors",
|
| 401 |
+
"language_model.model.layers.2.linear_attn.dt_bias": "model.safetensors",
|
| 402 |
+
"language_model.model.layers.2.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 403 |
+
"language_model.model.layers.2.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 404 |
+
"language_model.model.layers.2.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 405 |
+
"language_model.model.layers.2.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 406 |
+
"language_model.model.layers.2.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 407 |
+
"language_model.model.layers.2.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 408 |
+
"language_model.model.layers.2.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 409 |
+
"language_model.model.layers.2.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 410 |
+
"language_model.model.layers.2.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 411 |
+
"language_model.model.layers.2.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 412 |
+
"language_model.model.layers.2.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 413 |
+
"language_model.model.layers.2.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 414 |
+
"language_model.model.layers.2.linear_attn.norm.weight": "model.safetensors",
|
| 415 |
+
"language_model.model.layers.2.linear_attn.out_proj.biases": "model.safetensors",
|
| 416 |
+
"language_model.model.layers.2.linear_attn.out_proj.scales": "model.safetensors",
|
| 417 |
+
"language_model.model.layers.2.linear_attn.out_proj.weight": "model.safetensors",
|
| 418 |
+
"language_model.model.layers.2.mlp.down_proj.biases": "model.safetensors",
|
| 419 |
+
"language_model.model.layers.2.mlp.down_proj.scales": "model.safetensors",
|
| 420 |
+
"language_model.model.layers.2.mlp.down_proj.weight": "model.safetensors",
|
| 421 |
+
"language_model.model.layers.2.mlp.gate_proj.biases": "model.safetensors",
|
| 422 |
+
"language_model.model.layers.2.mlp.gate_proj.scales": "model.safetensors",
|
| 423 |
+
"language_model.model.layers.2.mlp.gate_proj.weight": "model.safetensors",
|
| 424 |
+
"language_model.model.layers.2.mlp.up_proj.biases": "model.safetensors",
|
| 425 |
+
"language_model.model.layers.2.mlp.up_proj.scales": "model.safetensors",
|
| 426 |
+
"language_model.model.layers.2.mlp.up_proj.weight": "model.safetensors",
|
| 427 |
+
"language_model.model.layers.2.post_attention_layernorm.weight": "model.safetensors",
|
| 428 |
+
"language_model.model.layers.20.input_layernorm.weight": "model.safetensors",
|
| 429 |
+
"language_model.model.layers.20.linear_attn.A_log": "model.safetensors",
|
| 430 |
+
"language_model.model.layers.20.linear_attn.conv1d.weight": "model.safetensors",
|
| 431 |
+
"language_model.model.layers.20.linear_attn.dt_bias": "model.safetensors",
|
| 432 |
+
"language_model.model.layers.20.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 433 |
+
"language_model.model.layers.20.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 434 |
+
"language_model.model.layers.20.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 435 |
+
"language_model.model.layers.20.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 436 |
+
"language_model.model.layers.20.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 437 |
+
"language_model.model.layers.20.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 438 |
+
"language_model.model.layers.20.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 439 |
+
"language_model.model.layers.20.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 440 |
+
"language_model.model.layers.20.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 441 |
+
"language_model.model.layers.20.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 442 |
+
"language_model.model.layers.20.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 443 |
+
"language_model.model.layers.20.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 444 |
+
"language_model.model.layers.20.linear_attn.norm.weight": "model.safetensors",
|
| 445 |
+
"language_model.model.layers.20.linear_attn.out_proj.biases": "model.safetensors",
|
| 446 |
+
"language_model.model.layers.20.linear_attn.out_proj.scales": "model.safetensors",
|
| 447 |
+
"language_model.model.layers.20.linear_attn.out_proj.weight": "model.safetensors",
|
| 448 |
+
"language_model.model.layers.20.mhc_attn.bias": "model.safetensors",
|
| 449 |
+
"language_model.model.layers.20.mhc_attn.weight": "model.safetensors",
|
| 450 |
+
"language_model.model.layers.20.mhc_ffn.bias": "model.safetensors",
|
| 451 |
+
"language_model.model.layers.20.mhc_ffn.weight": "model.safetensors",
|
| 452 |
+
"language_model.model.layers.20.mlp.down_proj.biases": "model.safetensors",
|
| 453 |
+
"language_model.model.layers.20.mlp.down_proj.scales": "model.safetensors",
|
| 454 |
+
"language_model.model.layers.20.mlp.down_proj.weight": "model.safetensors",
|
| 455 |
+
"language_model.model.layers.20.mlp.gate_proj.biases": "model.safetensors",
|
| 456 |
+
"language_model.model.layers.20.mlp.gate_proj.scales": "model.safetensors",
|
| 457 |
+
"language_model.model.layers.20.mlp.gate_proj.weight": "model.safetensors",
|
| 458 |
+
"language_model.model.layers.20.mlp.up_proj.biases": "model.safetensors",
|
| 459 |
+
"language_model.model.layers.20.mlp.up_proj.scales": "model.safetensors",
|
| 460 |
+
"language_model.model.layers.20.mlp.up_proj.weight": "model.safetensors",
|
| 461 |
+
"language_model.model.layers.20.post_attention_layernorm.weight": "model.safetensors",
|
| 462 |
+
"language_model.model.layers.21.input_layernorm.weight": "model.safetensors",
|
| 463 |
+
"language_model.model.layers.21.linear_attn.A_log": "model.safetensors",
|
| 464 |
+
"language_model.model.layers.21.linear_attn.conv1d.weight": "model.safetensors",
|
| 465 |
+
"language_model.model.layers.21.linear_attn.dt_bias": "model.safetensors",
|
| 466 |
+
"language_model.model.layers.21.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 467 |
+
"language_model.model.layers.21.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 468 |
+
"language_model.model.layers.21.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 469 |
+
"language_model.model.layers.21.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 470 |
+
"language_model.model.layers.21.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 471 |
+
"language_model.model.layers.21.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 472 |
+
"language_model.model.layers.21.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 473 |
+
"language_model.model.layers.21.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 474 |
+
"language_model.model.layers.21.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 475 |
+
"language_model.model.layers.21.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 476 |
+
"language_model.model.layers.21.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 477 |
+
"language_model.model.layers.21.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 478 |
+
"language_model.model.layers.21.linear_attn.norm.weight": "model.safetensors",
|
| 479 |
+
"language_model.model.layers.21.linear_attn.out_proj.biases": "model.safetensors",
|
| 480 |
+
"language_model.model.layers.21.linear_attn.out_proj.scales": "model.safetensors",
|
| 481 |
+
"language_model.model.layers.21.linear_attn.out_proj.weight": "model.safetensors",
|
| 482 |
+
"language_model.model.layers.21.mhc_attn.bias": "model.safetensors",
|
| 483 |
+
"language_model.model.layers.21.mhc_attn.weight": "model.safetensors",
|
| 484 |
+
"language_model.model.layers.21.mhc_ffn.bias": "model.safetensors",
|
| 485 |
+
"language_model.model.layers.21.mhc_ffn.weight": "model.safetensors",
|
| 486 |
+
"language_model.model.layers.21.mlp.down_proj.biases": "model.safetensors",
|
| 487 |
+
"language_model.model.layers.21.mlp.down_proj.scales": "model.safetensors",
|
| 488 |
+
"language_model.model.layers.21.mlp.down_proj.weight": "model.safetensors",
|
| 489 |
+
"language_model.model.layers.21.mlp.gate_proj.biases": "model.safetensors",
|
| 490 |
+
"language_model.model.layers.21.mlp.gate_proj.scales": "model.safetensors",
|
| 491 |
+
"language_model.model.layers.21.mlp.gate_proj.weight": "model.safetensors",
|
| 492 |
+
"language_model.model.layers.21.mlp.up_proj.biases": "model.safetensors",
|
| 493 |
+
"language_model.model.layers.21.mlp.up_proj.scales": "model.safetensors",
|
| 494 |
+
"language_model.model.layers.21.mlp.up_proj.weight": "model.safetensors",
|
| 495 |
+
"language_model.model.layers.21.post_attention_layernorm.weight": "model.safetensors",
|
| 496 |
+
"language_model.model.layers.22.input_layernorm.weight": "model.safetensors",
|
| 497 |
+
"language_model.model.layers.22.linear_attn.A_log": "model.safetensors",
|
| 498 |
+
"language_model.model.layers.22.linear_attn.conv1d.weight": "model.safetensors",
|
| 499 |
+
"language_model.model.layers.22.linear_attn.dt_bias": "model.safetensors",
|
| 500 |
+
"language_model.model.layers.22.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 501 |
+
"language_model.model.layers.22.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 502 |
+
"language_model.model.layers.22.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 503 |
+
"language_model.model.layers.22.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 504 |
+
"language_model.model.layers.22.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 505 |
+
"language_model.model.layers.22.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 506 |
+
"language_model.model.layers.22.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 507 |
+
"language_model.model.layers.22.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 508 |
+
"language_model.model.layers.22.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 509 |
+
"language_model.model.layers.22.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 510 |
+
"language_model.model.layers.22.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 511 |
+
"language_model.model.layers.22.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 512 |
+
"language_model.model.layers.22.linear_attn.norm.weight": "model.safetensors",
|
| 513 |
+
"language_model.model.layers.22.linear_attn.out_proj.biases": "model.safetensors",
|
| 514 |
+
"language_model.model.layers.22.linear_attn.out_proj.scales": "model.safetensors",
|
| 515 |
+
"language_model.model.layers.22.linear_attn.out_proj.weight": "model.safetensors",
|
| 516 |
+
"language_model.model.layers.22.mhc_attn.bias": "model.safetensors",
|
| 517 |
+
"language_model.model.layers.22.mhc_attn.weight": "model.safetensors",
|
| 518 |
+
"language_model.model.layers.22.mhc_ffn.bias": "model.safetensors",
|
| 519 |
+
"language_model.model.layers.22.mhc_ffn.weight": "model.safetensors",
|
| 520 |
+
"language_model.model.layers.22.mlp.down_proj.biases": "model.safetensors",
|
| 521 |
+
"language_model.model.layers.22.mlp.down_proj.scales": "model.safetensors",
|
| 522 |
+
"language_model.model.layers.22.mlp.down_proj.weight": "model.safetensors",
|
| 523 |
+
"language_model.model.layers.22.mlp.gate_proj.biases": "model.safetensors",
|
| 524 |
+
"language_model.model.layers.22.mlp.gate_proj.scales": "model.safetensors",
|
| 525 |
+
"language_model.model.layers.22.mlp.gate_proj.weight": "model.safetensors",
|
| 526 |
+
"language_model.model.layers.22.mlp.up_proj.biases": "model.safetensors",
|
| 527 |
+
"language_model.model.layers.22.mlp.up_proj.scales": "model.safetensors",
|
| 528 |
+
"language_model.model.layers.22.mlp.up_proj.weight": "model.safetensors",
|
| 529 |
+
"language_model.model.layers.22.post_attention_layernorm.weight": "model.safetensors",
|
| 530 |
+
"language_model.model.layers.23.input_layernorm.weight": "model.safetensors",
|
| 531 |
+
"language_model.model.layers.23.mhc_attn.bias": "model.safetensors",
|
| 532 |
+
"language_model.model.layers.23.mhc_attn.weight": "model.safetensors",
|
| 533 |
+
"language_model.model.layers.23.mhc_ffn.bias": "model.safetensors",
|
| 534 |
+
"language_model.model.layers.23.mhc_ffn.weight": "model.safetensors",
|
| 535 |
+
"language_model.model.layers.23.mlp.down_proj.biases": "model.safetensors",
|
| 536 |
+
"language_model.model.layers.23.mlp.down_proj.scales": "model.safetensors",
|
| 537 |
+
"language_model.model.layers.23.mlp.down_proj.weight": "model.safetensors",
|
| 538 |
+
"language_model.model.layers.23.mlp.gate_proj.biases": "model.safetensors",
|
| 539 |
+
"language_model.model.layers.23.mlp.gate_proj.scales": "model.safetensors",
|
| 540 |
+
"language_model.model.layers.23.mlp.gate_proj.weight": "model.safetensors",
|
| 541 |
+
"language_model.model.layers.23.mlp.up_proj.biases": "model.safetensors",
|
| 542 |
+
"language_model.model.layers.23.mlp.up_proj.scales": "model.safetensors",
|
| 543 |
+
"language_model.model.layers.23.mlp.up_proj.weight": "model.safetensors",
|
| 544 |
+
"language_model.model.layers.23.post_attention_layernorm.weight": "model.safetensors",
|
| 545 |
+
"language_model.model.layers.23.self_attn.k_norm.weight": "model.safetensors",
|
| 546 |
+
"language_model.model.layers.23.self_attn.k_proj.biases": "model.safetensors",
|
| 547 |
+
"language_model.model.layers.23.self_attn.k_proj.scales": "model.safetensors",
|
| 548 |
+
"language_model.model.layers.23.self_attn.k_proj.weight": "model.safetensors",
|
| 549 |
+
"language_model.model.layers.23.self_attn.o_proj.biases": "model.safetensors",
|
| 550 |
+
"language_model.model.layers.23.self_attn.o_proj.scales": "model.safetensors",
|
| 551 |
+
"language_model.model.layers.23.self_attn.o_proj.weight": "model.safetensors",
|
| 552 |
+
"language_model.model.layers.23.self_attn.q_norm.weight": "model.safetensors",
|
| 553 |
+
"language_model.model.layers.23.self_attn.q_proj.biases": "model.safetensors",
|
| 554 |
+
"language_model.model.layers.23.self_attn.q_proj.scales": "model.safetensors",
|
| 555 |
+
"language_model.model.layers.23.self_attn.q_proj.weight": "model.safetensors",
|
| 556 |
+
"language_model.model.layers.23.self_attn.v_proj.biases": "model.safetensors",
|
| 557 |
+
"language_model.model.layers.23.self_attn.v_proj.scales": "model.safetensors",
|
| 558 |
+
"language_model.model.layers.23.self_attn.v_proj.weight": "model.safetensors",
|
| 559 |
+
"language_model.model.layers.24.input_layernorm.weight": "model.safetensors",
|
| 560 |
+
"language_model.model.layers.24.linear_attn.A_log": "model.safetensors",
|
| 561 |
+
"language_model.model.layers.24.linear_attn.conv1d.weight": "model.safetensors",
|
| 562 |
+
"language_model.model.layers.24.linear_attn.dt_bias": "model.safetensors",
|
| 563 |
+
"language_model.model.layers.24.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 564 |
+
"language_model.model.layers.24.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 565 |
+
"language_model.model.layers.24.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 566 |
+
"language_model.model.layers.24.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 567 |
+
"language_model.model.layers.24.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 568 |
+
"language_model.model.layers.24.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 569 |
+
"language_model.model.layers.24.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 570 |
+
"language_model.model.layers.24.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 571 |
+
"language_model.model.layers.24.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 572 |
+
"language_model.model.layers.24.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 573 |
+
"language_model.model.layers.24.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 574 |
+
"language_model.model.layers.24.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 575 |
+
"language_model.model.layers.24.linear_attn.norm.weight": "model.safetensors",
|
| 576 |
+
"language_model.model.layers.24.linear_attn.out_proj.biases": "model.safetensors",
|
| 577 |
+
"language_model.model.layers.24.linear_attn.out_proj.scales": "model.safetensors",
|
| 578 |
+
"language_model.model.layers.24.linear_attn.out_proj.weight": "model.safetensors",
|
| 579 |
+
"language_model.model.layers.24.mhc_attn.bias": "model.safetensors",
|
| 580 |
+
"language_model.model.layers.24.mhc_attn.weight": "model.safetensors",
|
| 581 |
+
"language_model.model.layers.24.mhc_ffn.bias": "model.safetensors",
|
| 582 |
+
"language_model.model.layers.24.mhc_ffn.weight": "model.safetensors",
|
| 583 |
+
"language_model.model.layers.24.mlp.down_proj.biases": "model.safetensors",
|
| 584 |
+
"language_model.model.layers.24.mlp.down_proj.scales": "model.safetensors",
|
| 585 |
+
"language_model.model.layers.24.mlp.down_proj.weight": "model.safetensors",
|
| 586 |
+
"language_model.model.layers.24.mlp.gate_proj.biases": "model.safetensors",
|
| 587 |
+
"language_model.model.layers.24.mlp.gate_proj.scales": "model.safetensors",
|
| 588 |
+
"language_model.model.layers.24.mlp.gate_proj.weight": "model.safetensors",
|
| 589 |
+
"language_model.model.layers.24.mlp.up_proj.biases": "model.safetensors",
|
| 590 |
+
"language_model.model.layers.24.mlp.up_proj.scales": "model.safetensors",
|
| 591 |
+
"language_model.model.layers.24.mlp.up_proj.weight": "model.safetensors",
|
| 592 |
+
"language_model.model.layers.24.post_attention_layernorm.weight": "model.safetensors",
|
| 593 |
+
"language_model.model.layers.25.input_layernorm.weight": "model.safetensors",
|
| 594 |
+
"language_model.model.layers.25.linear_attn.A_log": "model.safetensors",
|
| 595 |
+
"language_model.model.layers.25.linear_attn.conv1d.weight": "model.safetensors",
|
| 596 |
+
"language_model.model.layers.25.linear_attn.dt_bias": "model.safetensors",
|
| 597 |
+
"language_model.model.layers.25.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 598 |
+
"language_model.model.layers.25.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 599 |
+
"language_model.model.layers.25.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 600 |
+
"language_model.model.layers.25.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 601 |
+
"language_model.model.layers.25.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 602 |
+
"language_model.model.layers.25.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 603 |
+
"language_model.model.layers.25.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 604 |
+
"language_model.model.layers.25.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 605 |
+
"language_model.model.layers.25.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 606 |
+
"language_model.model.layers.25.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 607 |
+
"language_model.model.layers.25.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 608 |
+
"language_model.model.layers.25.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 609 |
+
"language_model.model.layers.25.linear_attn.norm.weight": "model.safetensors",
|
| 610 |
+
"language_model.model.layers.25.linear_attn.out_proj.biases": "model.safetensors",
|
| 611 |
+
"language_model.model.layers.25.linear_attn.out_proj.scales": "model.safetensors",
|
| 612 |
+
"language_model.model.layers.25.linear_attn.out_proj.weight": "model.safetensors",
|
| 613 |
+
"language_model.model.layers.25.mhc_attn.bias": "model.safetensors",
|
| 614 |
+
"language_model.model.layers.25.mhc_attn.weight": "model.safetensors",
|
| 615 |
+
"language_model.model.layers.25.mhc_ffn.bias": "model.safetensors",
|
| 616 |
+
"language_model.model.layers.25.mhc_ffn.weight": "model.safetensors",
|
| 617 |
+
"language_model.model.layers.25.mlp.down_proj.biases": "model.safetensors",
|
| 618 |
+
"language_model.model.layers.25.mlp.down_proj.scales": "model.safetensors",
|
| 619 |
+
"language_model.model.layers.25.mlp.down_proj.weight": "model.safetensors",
|
| 620 |
+
"language_model.model.layers.25.mlp.gate_proj.biases": "model.safetensors",
|
| 621 |
+
"language_model.model.layers.25.mlp.gate_proj.scales": "model.safetensors",
|
| 622 |
+
"language_model.model.layers.25.mlp.gate_proj.weight": "model.safetensors",
|
| 623 |
+
"language_model.model.layers.25.mlp.up_proj.biases": "model.safetensors",
|
| 624 |
+
"language_model.model.layers.25.mlp.up_proj.scales": "model.safetensors",
|
| 625 |
+
"language_model.model.layers.25.mlp.up_proj.weight": "model.safetensors",
|
| 626 |
+
"language_model.model.layers.25.post_attention_layernorm.weight": "model.safetensors",
|
| 627 |
+
"language_model.model.layers.26.input_layernorm.weight": "model.safetensors",
|
| 628 |
+
"language_model.model.layers.26.linear_attn.A_log": "model.safetensors",
|
| 629 |
+
"language_model.model.layers.26.linear_attn.conv1d.weight": "model.safetensors",
|
| 630 |
+
"language_model.model.layers.26.linear_attn.dt_bias": "model.safetensors",
|
| 631 |
+
"language_model.model.layers.26.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 632 |
+
"language_model.model.layers.26.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 633 |
+
"language_model.model.layers.26.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 634 |
+
"language_model.model.layers.26.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 635 |
+
"language_model.model.layers.26.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 636 |
+
"language_model.model.layers.26.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 637 |
+
"language_model.model.layers.26.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 638 |
+
"language_model.model.layers.26.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 639 |
+
"language_model.model.layers.26.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 640 |
+
"language_model.model.layers.26.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 641 |
+
"language_model.model.layers.26.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 642 |
+
"language_model.model.layers.26.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 643 |
+
"language_model.model.layers.26.linear_attn.norm.weight": "model.safetensors",
|
| 644 |
+
"language_model.model.layers.26.linear_attn.out_proj.biases": "model.safetensors",
|
| 645 |
+
"language_model.model.layers.26.linear_attn.out_proj.scales": "model.safetensors",
|
| 646 |
+
"language_model.model.layers.26.linear_attn.out_proj.weight": "model.safetensors",
|
| 647 |
+
"language_model.model.layers.26.mhc_attn.bias": "model.safetensors",
|
| 648 |
+
"language_model.model.layers.26.mhc_attn.weight": "model.safetensors",
|
| 649 |
+
"language_model.model.layers.26.mhc_ffn.bias": "model.safetensors",
|
| 650 |
+
"language_model.model.layers.26.mhc_ffn.weight": "model.safetensors",
|
| 651 |
+
"language_model.model.layers.26.mlp.down_proj.biases": "model.safetensors",
|
| 652 |
+
"language_model.model.layers.26.mlp.down_proj.scales": "model.safetensors",
|
| 653 |
+
"language_model.model.layers.26.mlp.down_proj.weight": "model.safetensors",
|
| 654 |
+
"language_model.model.layers.26.mlp.gate_proj.biases": "model.safetensors",
|
| 655 |
+
"language_model.model.layers.26.mlp.gate_proj.scales": "model.safetensors",
|
| 656 |
+
"language_model.model.layers.26.mlp.gate_proj.weight": "model.safetensors",
|
| 657 |
+
"language_model.model.layers.26.mlp.up_proj.biases": "model.safetensors",
|
| 658 |
+
"language_model.model.layers.26.mlp.up_proj.scales": "model.safetensors",
|
| 659 |
+
"language_model.model.layers.26.mlp.up_proj.weight": "model.safetensors",
|
| 660 |
+
"language_model.model.layers.26.post_attention_layernorm.weight": "model.safetensors",
|
| 661 |
+
"language_model.model.layers.27.input_layernorm.weight": "model.safetensors",
|
| 662 |
+
"language_model.model.layers.27.mhc_attn.bias": "model.safetensors",
|
| 663 |
+
"language_model.model.layers.27.mhc_attn.weight": "model.safetensors",
|
| 664 |
+
"language_model.model.layers.27.mhc_ffn.bias": "model.safetensors",
|
| 665 |
+
"language_model.model.layers.27.mhc_ffn.weight": "model.safetensors",
|
| 666 |
+
"language_model.model.layers.27.mlp.down_proj.biases": "model.safetensors",
|
| 667 |
+
"language_model.model.layers.27.mlp.down_proj.scales": "model.safetensors",
|
| 668 |
+
"language_model.model.layers.27.mlp.down_proj.weight": "model.safetensors",
|
| 669 |
+
"language_model.model.layers.27.mlp.gate_proj.biases": "model.safetensors",
|
| 670 |
+
"language_model.model.layers.27.mlp.gate_proj.scales": "model.safetensors",
|
| 671 |
+
"language_model.model.layers.27.mlp.gate_proj.weight": "model.safetensors",
|
| 672 |
+
"language_model.model.layers.27.mlp.up_proj.biases": "model.safetensors",
|
| 673 |
+
"language_model.model.layers.27.mlp.up_proj.scales": "model.safetensors",
|
| 674 |
+
"language_model.model.layers.27.mlp.up_proj.weight": "model.safetensors",
|
| 675 |
+
"language_model.model.layers.27.post_attention_layernorm.weight": "model.safetensors",
|
| 676 |
+
"language_model.model.layers.27.self_attn.k_norm.weight": "model.safetensors",
|
| 677 |
+
"language_model.model.layers.27.self_attn.k_proj.biases": "model.safetensors",
|
| 678 |
+
"language_model.model.layers.27.self_attn.k_proj.scales": "model.safetensors",
|
| 679 |
+
"language_model.model.layers.27.self_attn.k_proj.weight": "model.safetensors",
|
| 680 |
+
"language_model.model.layers.27.self_attn.o_proj.biases": "model.safetensors",
|
| 681 |
+
"language_model.model.layers.27.self_attn.o_proj.scales": "model.safetensors",
|
| 682 |
+
"language_model.model.layers.27.self_attn.o_proj.weight": "model.safetensors",
|
| 683 |
+
"language_model.model.layers.27.self_attn.q_norm.weight": "model.safetensors",
|
| 684 |
+
"language_model.model.layers.27.self_attn.q_proj.biases": "model.safetensors",
|
| 685 |
+
"language_model.model.layers.27.self_attn.q_proj.scales": "model.safetensors",
|
| 686 |
+
"language_model.model.layers.27.self_attn.q_proj.weight": "model.safetensors",
|
| 687 |
+
"language_model.model.layers.27.self_attn.v_proj.biases": "model.safetensors",
|
| 688 |
+
"language_model.model.layers.27.self_attn.v_proj.scales": "model.safetensors",
|
| 689 |
+
"language_model.model.layers.27.self_attn.v_proj.weight": "model.safetensors",
|
| 690 |
+
"language_model.model.layers.28.input_layernorm.weight": "model.safetensors",
|
| 691 |
+
"language_model.model.layers.28.linear_attn.A_log": "model.safetensors",
|
| 692 |
+
"language_model.model.layers.28.linear_attn.conv1d.weight": "model.safetensors",
|
| 693 |
+
"language_model.model.layers.28.linear_attn.dt_bias": "model.safetensors",
|
| 694 |
+
"language_model.model.layers.28.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 695 |
+
"language_model.model.layers.28.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 696 |
+
"language_model.model.layers.28.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 697 |
+
"language_model.model.layers.28.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 698 |
+
"language_model.model.layers.28.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 699 |
+
"language_model.model.layers.28.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 700 |
+
"language_model.model.layers.28.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 701 |
+
"language_model.model.layers.28.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 702 |
+
"language_model.model.layers.28.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 703 |
+
"language_model.model.layers.28.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 704 |
+
"language_model.model.layers.28.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 705 |
+
"language_model.model.layers.28.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 706 |
+
"language_model.model.layers.28.linear_attn.norm.weight": "model.safetensors",
|
| 707 |
+
"language_model.model.layers.28.linear_attn.out_proj.biases": "model.safetensors",
|
| 708 |
+
"language_model.model.layers.28.linear_attn.out_proj.scales": "model.safetensors",
|
| 709 |
+
"language_model.model.layers.28.linear_attn.out_proj.weight": "model.safetensors",
|
| 710 |
+
"language_model.model.layers.28.mlp.down_proj.biases": "model.safetensors",
|
| 711 |
+
"language_model.model.layers.28.mlp.down_proj.scales": "model.safetensors",
|
| 712 |
+
"language_model.model.layers.28.mlp.down_proj.weight": "model.safetensors",
|
| 713 |
+
"language_model.model.layers.28.mlp.gate_proj.biases": "model.safetensors",
|
| 714 |
+
"language_model.model.layers.28.mlp.gate_proj.scales": "model.safetensors",
|
| 715 |
+
"language_model.model.layers.28.mlp.gate_proj.weight": "model.safetensors",
|
| 716 |
+
"language_model.model.layers.28.mlp.up_proj.biases": "model.safetensors",
|
| 717 |
+
"language_model.model.layers.28.mlp.up_proj.scales": "model.safetensors",
|
| 718 |
+
"language_model.model.layers.28.mlp.up_proj.weight": "model.safetensors",
|
| 719 |
+
"language_model.model.layers.28.post_attention_layernorm.weight": "model.safetensors",
|
| 720 |
+
"language_model.model.layers.29.input_layernorm.weight": "model.safetensors",
|
| 721 |
+
"language_model.model.layers.29.linear_attn.A_log": "model.safetensors",
|
| 722 |
+
"language_model.model.layers.29.linear_attn.conv1d.weight": "model.safetensors",
|
| 723 |
+
"language_model.model.layers.29.linear_attn.dt_bias": "model.safetensors",
|
| 724 |
+
"language_model.model.layers.29.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 725 |
+
"language_model.model.layers.29.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 726 |
+
"language_model.model.layers.29.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 727 |
+
"language_model.model.layers.29.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 728 |
+
"language_model.model.layers.29.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 729 |
+
"language_model.model.layers.29.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 730 |
+
"language_model.model.layers.29.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 731 |
+
"language_model.model.layers.29.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 732 |
+
"language_model.model.layers.29.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 733 |
+
"language_model.model.layers.29.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 734 |
+
"language_model.model.layers.29.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 735 |
+
"language_model.model.layers.29.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 736 |
+
"language_model.model.layers.29.linear_attn.norm.weight": "model.safetensors",
|
| 737 |
+
"language_model.model.layers.29.linear_attn.out_proj.biases": "model.safetensors",
|
| 738 |
+
"language_model.model.layers.29.linear_attn.out_proj.scales": "model.safetensors",
|
| 739 |
+
"language_model.model.layers.29.linear_attn.out_proj.weight": "model.safetensors",
|
| 740 |
+
"language_model.model.layers.29.mlp.down_proj.biases": "model.safetensors",
|
| 741 |
+
"language_model.model.layers.29.mlp.down_proj.scales": "model.safetensors",
|
| 742 |
+
"language_model.model.layers.29.mlp.down_proj.weight": "model.safetensors",
|
| 743 |
+
"language_model.model.layers.29.mlp.gate_proj.biases": "model.safetensors",
|
| 744 |
+
"language_model.model.layers.29.mlp.gate_proj.scales": "model.safetensors",
|
| 745 |
+
"language_model.model.layers.29.mlp.gate_proj.weight": "model.safetensors",
|
| 746 |
+
"language_model.model.layers.29.mlp.up_proj.biases": "model.safetensors",
|
| 747 |
+
"language_model.model.layers.29.mlp.up_proj.scales": "model.safetensors",
|
| 748 |
+
"language_model.model.layers.29.mlp.up_proj.weight": "model.safetensors",
|
| 749 |
+
"language_model.model.layers.29.post_attention_layernorm.weight": "model.safetensors",
|
| 750 |
+
"language_model.model.layers.3.input_layernorm.weight": "model.safetensors",
|
| 751 |
+
"language_model.model.layers.3.mlp.down_proj.biases": "model.safetensors",
|
| 752 |
+
"language_model.model.layers.3.mlp.down_proj.scales": "model.safetensors",
|
| 753 |
+
"language_model.model.layers.3.mlp.down_proj.weight": "model.safetensors",
|
| 754 |
+
"language_model.model.layers.3.mlp.gate_proj.biases": "model.safetensors",
|
| 755 |
+
"language_model.model.layers.3.mlp.gate_proj.scales": "model.safetensors",
|
| 756 |
+
"language_model.model.layers.3.mlp.gate_proj.weight": "model.safetensors",
|
| 757 |
+
"language_model.model.layers.3.mlp.up_proj.biases": "model.safetensors",
|
| 758 |
+
"language_model.model.layers.3.mlp.up_proj.scales": "model.safetensors",
|
| 759 |
+
"language_model.model.layers.3.mlp.up_proj.weight": "model.safetensors",
|
| 760 |
+
"language_model.model.layers.3.post_attention_layernorm.weight": "model.safetensors",
|
| 761 |
+
"language_model.model.layers.3.self_attn.k_norm.weight": "model.safetensors",
|
| 762 |
+
"language_model.model.layers.3.self_attn.k_proj.biases": "model.safetensors",
|
| 763 |
+
"language_model.model.layers.3.self_attn.k_proj.scales": "model.safetensors",
|
| 764 |
+
"language_model.model.layers.3.self_attn.k_proj.weight": "model.safetensors",
|
| 765 |
+
"language_model.model.layers.3.self_attn.o_proj.biases": "model.safetensors",
|
| 766 |
+
"language_model.model.layers.3.self_attn.o_proj.scales": "model.safetensors",
|
| 767 |
+
"language_model.model.layers.3.self_attn.o_proj.weight": "model.safetensors",
|
| 768 |
+
"language_model.model.layers.3.self_attn.q_norm.weight": "model.safetensors",
|
| 769 |
+
"language_model.model.layers.3.self_attn.q_proj.biases": "model.safetensors",
|
| 770 |
+
"language_model.model.layers.3.self_attn.q_proj.scales": "model.safetensors",
|
| 771 |
+
"language_model.model.layers.3.self_attn.q_proj.weight": "model.safetensors",
|
| 772 |
+
"language_model.model.layers.3.self_attn.v_proj.biases": "model.safetensors",
|
| 773 |
+
"language_model.model.layers.3.self_attn.v_proj.scales": "model.safetensors",
|
| 774 |
+
"language_model.model.layers.3.self_attn.v_proj.weight": "model.safetensors",
|
| 775 |
+
"language_model.model.layers.30.input_layernorm.weight": "model.safetensors",
|
| 776 |
+
"language_model.model.layers.30.linear_attn.A_log": "model.safetensors",
|
| 777 |
+
"language_model.model.layers.30.linear_attn.conv1d.weight": "model.safetensors",
|
| 778 |
+
"language_model.model.layers.30.linear_attn.dt_bias": "model.safetensors",
|
| 779 |
+
"language_model.model.layers.30.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 780 |
+
"language_model.model.layers.30.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 781 |
+
"language_model.model.layers.30.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 782 |
+
"language_model.model.layers.30.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 783 |
+
"language_model.model.layers.30.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 784 |
+
"language_model.model.layers.30.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 785 |
+
"language_model.model.layers.30.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 786 |
+
"language_model.model.layers.30.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 787 |
+
"language_model.model.layers.30.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 788 |
+
"language_model.model.layers.30.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 789 |
+
"language_model.model.layers.30.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 790 |
+
"language_model.model.layers.30.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 791 |
+
"language_model.model.layers.30.linear_attn.norm.weight": "model.safetensors",
|
| 792 |
+
"language_model.model.layers.30.linear_attn.out_proj.biases": "model.safetensors",
|
| 793 |
+
"language_model.model.layers.30.linear_attn.out_proj.scales": "model.safetensors",
|
| 794 |
+
"language_model.model.layers.30.linear_attn.out_proj.weight": "model.safetensors",
|
| 795 |
+
"language_model.model.layers.30.mlp.down_proj.biases": "model.safetensors",
|
| 796 |
+
"language_model.model.layers.30.mlp.down_proj.scales": "model.safetensors",
|
| 797 |
+
"language_model.model.layers.30.mlp.down_proj.weight": "model.safetensors",
|
| 798 |
+
"language_model.model.layers.30.mlp.gate_proj.biases": "model.safetensors",
|
| 799 |
+
"language_model.model.layers.30.mlp.gate_proj.scales": "model.safetensors",
|
| 800 |
+
"language_model.model.layers.30.mlp.gate_proj.weight": "model.safetensors",
|
| 801 |
+
"language_model.model.layers.30.mlp.up_proj.biases": "model.safetensors",
|
| 802 |
+
"language_model.model.layers.30.mlp.up_proj.scales": "model.safetensors",
|
| 803 |
+
"language_model.model.layers.30.mlp.up_proj.weight": "model.safetensors",
|
| 804 |
+
"language_model.model.layers.30.post_attention_layernorm.weight": "model.safetensors",
|
| 805 |
+
"language_model.model.layers.31.input_layernorm.weight": "model.safetensors",
|
| 806 |
+
"language_model.model.layers.31.mlp.down_proj.biases": "model.safetensors",
|
| 807 |
+
"language_model.model.layers.31.mlp.down_proj.scales": "model.safetensors",
|
| 808 |
+
"language_model.model.layers.31.mlp.down_proj.weight": "model.safetensors",
|
| 809 |
+
"language_model.model.layers.31.mlp.gate_proj.biases": "model.safetensors",
|
| 810 |
+
"language_model.model.layers.31.mlp.gate_proj.scales": "model.safetensors",
|
| 811 |
+
"language_model.model.layers.31.mlp.gate_proj.weight": "model.safetensors",
|
| 812 |
+
"language_model.model.layers.31.mlp.up_proj.biases": "model.safetensors",
|
| 813 |
+
"language_model.model.layers.31.mlp.up_proj.scales": "model.safetensors",
|
| 814 |
+
"language_model.model.layers.31.mlp.up_proj.weight": "model.safetensors",
|
| 815 |
+
"language_model.model.layers.31.post_attention_layernorm.weight": "model.safetensors",
|
| 816 |
+
"language_model.model.layers.31.self_attn.k_norm.weight": "model.safetensors",
|
| 817 |
+
"language_model.model.layers.31.self_attn.k_proj.biases": "model.safetensors",
|
| 818 |
+
"language_model.model.layers.31.self_attn.k_proj.scales": "model.safetensors",
|
| 819 |
+
"language_model.model.layers.31.self_attn.k_proj.weight": "model.safetensors",
|
| 820 |
+
"language_model.model.layers.31.self_attn.o_proj.biases": "model.safetensors",
|
| 821 |
+
"language_model.model.layers.31.self_attn.o_proj.scales": "model.safetensors",
|
| 822 |
+
"language_model.model.layers.31.self_attn.o_proj.weight": "model.safetensors",
|
| 823 |
+
"language_model.model.layers.31.self_attn.q_norm.weight": "model.safetensors",
|
| 824 |
+
"language_model.model.layers.31.self_attn.q_proj.biases": "model.safetensors",
|
| 825 |
+
"language_model.model.layers.31.self_attn.q_proj.scales": "model.safetensors",
|
| 826 |
+
"language_model.model.layers.31.self_attn.q_proj.weight": "model.safetensors",
|
| 827 |
+
"language_model.model.layers.31.self_attn.v_proj.biases": "model.safetensors",
|
| 828 |
+
"language_model.model.layers.31.self_attn.v_proj.scales": "model.safetensors",
|
| 829 |
+
"language_model.model.layers.31.self_attn.v_proj.weight": "model.safetensors",
|
| 830 |
+
"language_model.model.layers.4.input_layernorm.weight": "model.safetensors",
|
| 831 |
+
"language_model.model.layers.4.linear_attn.A_log": "model.safetensors",
|
| 832 |
+
"language_model.model.layers.4.linear_attn.conv1d.weight": "model.safetensors",
|
| 833 |
+
"language_model.model.layers.4.linear_attn.dt_bias": "model.safetensors",
|
| 834 |
+
"language_model.model.layers.4.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 835 |
+
"language_model.model.layers.4.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 836 |
+
"language_model.model.layers.4.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 837 |
+
"language_model.model.layers.4.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 838 |
+
"language_model.model.layers.4.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 839 |
+
"language_model.model.layers.4.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 840 |
+
"language_model.model.layers.4.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 841 |
+
"language_model.model.layers.4.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 842 |
+
"language_model.model.layers.4.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 843 |
+
"language_model.model.layers.4.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 844 |
+
"language_model.model.layers.4.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 845 |
+
"language_model.model.layers.4.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 846 |
+
"language_model.model.layers.4.linear_attn.norm.weight": "model.safetensors",
|
| 847 |
+
"language_model.model.layers.4.linear_attn.out_proj.biases": "model.safetensors",
|
| 848 |
+
"language_model.model.layers.4.linear_attn.out_proj.scales": "model.safetensors",
|
| 849 |
+
"language_model.model.layers.4.linear_attn.out_proj.weight": "model.safetensors",
|
| 850 |
+
"language_model.model.layers.4.mhc_attn.bias": "model.safetensors",
|
| 851 |
+
"language_model.model.layers.4.mhc_attn.weight": "model.safetensors",
|
| 852 |
+
"language_model.model.layers.4.mhc_ffn.bias": "model.safetensors",
|
| 853 |
+
"language_model.model.layers.4.mhc_ffn.weight": "model.safetensors",
|
| 854 |
+
"language_model.model.layers.4.mlp.down_proj.biases": "model.safetensors",
|
| 855 |
+
"language_model.model.layers.4.mlp.down_proj.scales": "model.safetensors",
|
| 856 |
+
"language_model.model.layers.4.mlp.down_proj.weight": "model.safetensors",
|
| 857 |
+
"language_model.model.layers.4.mlp.gate_proj.biases": "model.safetensors",
|
| 858 |
+
"language_model.model.layers.4.mlp.gate_proj.scales": "model.safetensors",
|
| 859 |
+
"language_model.model.layers.4.mlp.gate_proj.weight": "model.safetensors",
|
| 860 |
+
"language_model.model.layers.4.mlp.up_proj.biases": "model.safetensors",
|
| 861 |
+
"language_model.model.layers.4.mlp.up_proj.scales": "model.safetensors",
|
| 862 |
+
"language_model.model.layers.4.mlp.up_proj.weight": "model.safetensors",
|
| 863 |
+
"language_model.model.layers.4.post_attention_layernorm.weight": "model.safetensors",
|
| 864 |
+
"language_model.model.layers.5.input_layernorm.weight": "model.safetensors",
|
| 865 |
+
"language_model.model.layers.5.linear_attn.A_log": "model.safetensors",
|
| 866 |
+
"language_model.model.layers.5.linear_attn.conv1d.weight": "model.safetensors",
|
| 867 |
+
"language_model.model.layers.5.linear_attn.dt_bias": "model.safetensors",
|
| 868 |
+
"language_model.model.layers.5.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 869 |
+
"language_model.model.layers.5.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 870 |
+
"language_model.model.layers.5.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 871 |
+
"language_model.model.layers.5.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 872 |
+
"language_model.model.layers.5.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 873 |
+
"language_model.model.layers.5.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 874 |
+
"language_model.model.layers.5.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 875 |
+
"language_model.model.layers.5.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 876 |
+
"language_model.model.layers.5.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 877 |
+
"language_model.model.layers.5.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 878 |
+
"language_model.model.layers.5.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 879 |
+
"language_model.model.layers.5.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 880 |
+
"language_model.model.layers.5.linear_attn.norm.weight": "model.safetensors",
|
| 881 |
+
"language_model.model.layers.5.linear_attn.out_proj.biases": "model.safetensors",
|
| 882 |
+
"language_model.model.layers.5.linear_attn.out_proj.scales": "model.safetensors",
|
| 883 |
+
"language_model.model.layers.5.linear_attn.out_proj.weight": "model.safetensors",
|
| 884 |
+
"language_model.model.layers.5.mhc_attn.bias": "model.safetensors",
|
| 885 |
+
"language_model.model.layers.5.mhc_attn.weight": "model.safetensors",
|
| 886 |
+
"language_model.model.layers.5.mhc_ffn.bias": "model.safetensors",
|
| 887 |
+
"language_model.model.layers.5.mhc_ffn.weight": "model.safetensors",
|
| 888 |
+
"language_model.model.layers.5.mlp.down_proj.biases": "model.safetensors",
|
| 889 |
+
"language_model.model.layers.5.mlp.down_proj.scales": "model.safetensors",
|
| 890 |
+
"language_model.model.layers.5.mlp.down_proj.weight": "model.safetensors",
|
| 891 |
+
"language_model.model.layers.5.mlp.gate_proj.biases": "model.safetensors",
|
| 892 |
+
"language_model.model.layers.5.mlp.gate_proj.scales": "model.safetensors",
|
| 893 |
+
"language_model.model.layers.5.mlp.gate_proj.weight": "model.safetensors",
|
| 894 |
+
"language_model.model.layers.5.mlp.up_proj.biases": "model.safetensors",
|
| 895 |
+
"language_model.model.layers.5.mlp.up_proj.scales": "model.safetensors",
|
| 896 |
+
"language_model.model.layers.5.mlp.up_proj.weight": "model.safetensors",
|
| 897 |
+
"language_model.model.layers.5.post_attention_layernorm.weight": "model.safetensors",
|
| 898 |
+
"language_model.model.layers.6.input_layernorm.weight": "model.safetensors",
|
| 899 |
+
"language_model.model.layers.6.linear_attn.A_log": "model.safetensors",
|
| 900 |
+
"language_model.model.layers.6.linear_attn.conv1d.weight": "model.safetensors",
|
| 901 |
+
"language_model.model.layers.6.linear_attn.dt_bias": "model.safetensors",
|
| 902 |
+
"language_model.model.layers.6.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 903 |
+
"language_model.model.layers.6.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 904 |
+
"language_model.model.layers.6.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 905 |
+
"language_model.model.layers.6.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 906 |
+
"language_model.model.layers.6.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 907 |
+
"language_model.model.layers.6.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 908 |
+
"language_model.model.layers.6.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 909 |
+
"language_model.model.layers.6.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 910 |
+
"language_model.model.layers.6.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 911 |
+
"language_model.model.layers.6.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 912 |
+
"language_model.model.layers.6.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 913 |
+
"language_model.model.layers.6.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 914 |
+
"language_model.model.layers.6.linear_attn.norm.weight": "model.safetensors",
|
| 915 |
+
"language_model.model.layers.6.linear_attn.out_proj.biases": "model.safetensors",
|
| 916 |
+
"language_model.model.layers.6.linear_attn.out_proj.scales": "model.safetensors",
|
| 917 |
+
"language_model.model.layers.6.linear_attn.out_proj.weight": "model.safetensors",
|
| 918 |
+
"language_model.model.layers.6.mhc_attn.bias": "model.safetensors",
|
| 919 |
+
"language_model.model.layers.6.mhc_attn.weight": "model.safetensors",
|
| 920 |
+
"language_model.model.layers.6.mhc_ffn.bias": "model.safetensors",
|
| 921 |
+
"language_model.model.layers.6.mhc_ffn.weight": "model.safetensors",
|
| 922 |
+
"language_model.model.layers.6.mlp.down_proj.biases": "model.safetensors",
|
| 923 |
+
"language_model.model.layers.6.mlp.down_proj.scales": "model.safetensors",
|
| 924 |
+
"language_model.model.layers.6.mlp.down_proj.weight": "model.safetensors",
|
| 925 |
+
"language_model.model.layers.6.mlp.gate_proj.biases": "model.safetensors",
|
| 926 |
+
"language_model.model.layers.6.mlp.gate_proj.scales": "model.safetensors",
|
| 927 |
+
"language_model.model.layers.6.mlp.gate_proj.weight": "model.safetensors",
|
| 928 |
+
"language_model.model.layers.6.mlp.up_proj.biases": "model.safetensors",
|
| 929 |
+
"language_model.model.layers.6.mlp.up_proj.scales": "model.safetensors",
|
| 930 |
+
"language_model.model.layers.6.mlp.up_proj.weight": "model.safetensors",
|
| 931 |
+
"language_model.model.layers.6.post_attention_layernorm.weight": "model.safetensors",
|
| 932 |
+
"language_model.model.layers.7.input_layernorm.weight": "model.safetensors",
|
| 933 |
+
"language_model.model.layers.7.mhc_attn.bias": "model.safetensors",
|
| 934 |
+
"language_model.model.layers.7.mhc_attn.weight": "model.safetensors",
|
| 935 |
+
"language_model.model.layers.7.mhc_ffn.bias": "model.safetensors",
|
| 936 |
+
"language_model.model.layers.7.mhc_ffn.weight": "model.safetensors",
|
| 937 |
+
"language_model.model.layers.7.mlp.down_proj.biases": "model.safetensors",
|
| 938 |
+
"language_model.model.layers.7.mlp.down_proj.scales": "model.safetensors",
|
| 939 |
+
"language_model.model.layers.7.mlp.down_proj.weight": "model.safetensors",
|
| 940 |
+
"language_model.model.layers.7.mlp.gate_proj.biases": "model.safetensors",
|
| 941 |
+
"language_model.model.layers.7.mlp.gate_proj.scales": "model.safetensors",
|
| 942 |
+
"language_model.model.layers.7.mlp.gate_proj.weight": "model.safetensors",
|
| 943 |
+
"language_model.model.layers.7.mlp.up_proj.biases": "model.safetensors",
|
| 944 |
+
"language_model.model.layers.7.mlp.up_proj.scales": "model.safetensors",
|
| 945 |
+
"language_model.model.layers.7.mlp.up_proj.weight": "model.safetensors",
|
| 946 |
+
"language_model.model.layers.7.post_attention_layernorm.weight": "model.safetensors",
|
| 947 |
+
"language_model.model.layers.7.self_attn.k_norm.weight": "model.safetensors",
|
| 948 |
+
"language_model.model.layers.7.self_attn.k_proj.biases": "model.safetensors",
|
| 949 |
+
"language_model.model.layers.7.self_attn.k_proj.scales": "model.safetensors",
|
| 950 |
+
"language_model.model.layers.7.self_attn.k_proj.weight": "model.safetensors",
|
| 951 |
+
"language_model.model.layers.7.self_attn.o_proj.biases": "model.safetensors",
|
| 952 |
+
"language_model.model.layers.7.self_attn.o_proj.scales": "model.safetensors",
|
| 953 |
+
"language_model.model.layers.7.self_attn.o_proj.weight": "model.safetensors",
|
| 954 |
+
"language_model.model.layers.7.self_attn.q_norm.weight": "model.safetensors",
|
| 955 |
+
"language_model.model.layers.7.self_attn.q_proj.biases": "model.safetensors",
|
| 956 |
+
"language_model.model.layers.7.self_attn.q_proj.scales": "model.safetensors",
|
| 957 |
+
"language_model.model.layers.7.self_attn.q_proj.weight": "model.safetensors",
|
| 958 |
+
"language_model.model.layers.7.self_attn.v_proj.biases": "model.safetensors",
|
| 959 |
+
"language_model.model.layers.7.self_attn.v_proj.scales": "model.safetensors",
|
| 960 |
+
"language_model.model.layers.7.self_attn.v_proj.weight": "model.safetensors",
|
| 961 |
+
"language_model.model.layers.8.input_layernorm.weight": "model.safetensors",
|
| 962 |
+
"language_model.model.layers.8.linear_attn.A_log": "model.safetensors",
|
| 963 |
+
"language_model.model.layers.8.linear_attn.conv1d.weight": "model.safetensors",
|
| 964 |
+
"language_model.model.layers.8.linear_attn.dt_bias": "model.safetensors",
|
| 965 |
+
"language_model.model.layers.8.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 966 |
+
"language_model.model.layers.8.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 967 |
+
"language_model.model.layers.8.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 968 |
+
"language_model.model.layers.8.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 969 |
+
"language_model.model.layers.8.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 970 |
+
"language_model.model.layers.8.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 971 |
+
"language_model.model.layers.8.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 972 |
+
"language_model.model.layers.8.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 973 |
+
"language_model.model.layers.8.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 974 |
+
"language_model.model.layers.8.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 975 |
+
"language_model.model.layers.8.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 976 |
+
"language_model.model.layers.8.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 977 |
+
"language_model.model.layers.8.linear_attn.norm.weight": "model.safetensors",
|
| 978 |
+
"language_model.model.layers.8.linear_attn.out_proj.biases": "model.safetensors",
|
| 979 |
+
"language_model.model.layers.8.linear_attn.out_proj.scales": "model.safetensors",
|
| 980 |
+
"language_model.model.layers.8.linear_attn.out_proj.weight": "model.safetensors",
|
| 981 |
+
"language_model.model.layers.8.mhc_attn.bias": "model.safetensors",
|
| 982 |
+
"language_model.model.layers.8.mhc_attn.weight": "model.safetensors",
|
| 983 |
+
"language_model.model.layers.8.mhc_ffn.bias": "model.safetensors",
|
| 984 |
+
"language_model.model.layers.8.mhc_ffn.weight": "model.safetensors",
|
| 985 |
+
"language_model.model.layers.8.mlp.down_proj.biases": "model.safetensors",
|
| 986 |
+
"language_model.model.layers.8.mlp.down_proj.scales": "model.safetensors",
|
| 987 |
+
"language_model.model.layers.8.mlp.down_proj.weight": "model.safetensors",
|
| 988 |
+
"language_model.model.layers.8.mlp.gate_proj.biases": "model.safetensors",
|
| 989 |
+
"language_model.model.layers.8.mlp.gate_proj.scales": "model.safetensors",
|
| 990 |
+
"language_model.model.layers.8.mlp.gate_proj.weight": "model.safetensors",
|
| 991 |
+
"language_model.model.layers.8.mlp.up_proj.biases": "model.safetensors",
|
| 992 |
+
"language_model.model.layers.8.mlp.up_proj.scales": "model.safetensors",
|
| 993 |
+
"language_model.model.layers.8.mlp.up_proj.weight": "model.safetensors",
|
| 994 |
+
"language_model.model.layers.8.post_attention_layernorm.weight": "model.safetensors",
|
| 995 |
+
"language_model.model.layers.9.input_layernorm.weight": "model.safetensors",
|
| 996 |
+
"language_model.model.layers.9.linear_attn.A_log": "model.safetensors",
|
| 997 |
+
"language_model.model.layers.9.linear_attn.conv1d.weight": "model.safetensors",
|
| 998 |
+
"language_model.model.layers.9.linear_attn.dt_bias": "model.safetensors",
|
| 999 |
+
"language_model.model.layers.9.linear_attn.in_proj_a.biases": "model.safetensors",
|
| 1000 |
+
"language_model.model.layers.9.linear_attn.in_proj_a.scales": "model.safetensors",
|
| 1001 |
+
"language_model.model.layers.9.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 1002 |
+
"language_model.model.layers.9.linear_attn.in_proj_b.biases": "model.safetensors",
|
| 1003 |
+
"language_model.model.layers.9.linear_attn.in_proj_b.scales": "model.safetensors",
|
| 1004 |
+
"language_model.model.layers.9.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 1005 |
+
"language_model.model.layers.9.linear_attn.in_proj_qkv.biases": "model.safetensors",
|
| 1006 |
+
"language_model.model.layers.9.linear_attn.in_proj_qkv.scales": "model.safetensors",
|
| 1007 |
+
"language_model.model.layers.9.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 1008 |
+
"language_model.model.layers.9.linear_attn.in_proj_z.biases": "model.safetensors",
|
| 1009 |
+
"language_model.model.layers.9.linear_attn.in_proj_z.scales": "model.safetensors",
|
| 1010 |
+
"language_model.model.layers.9.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 1011 |
+
"language_model.model.layers.9.linear_attn.norm.weight": "model.safetensors",
|
| 1012 |
+
"language_model.model.layers.9.linear_attn.out_proj.biases": "model.safetensors",
|
| 1013 |
+
"language_model.model.layers.9.linear_attn.out_proj.scales": "model.safetensors",
|
| 1014 |
+
"language_model.model.layers.9.linear_attn.out_proj.weight": "model.safetensors",
|
| 1015 |
+
"language_model.model.layers.9.mhc_attn.bias": "model.safetensors",
|
| 1016 |
+
"language_model.model.layers.9.mhc_attn.weight": "model.safetensors",
|
| 1017 |
+
"language_model.model.layers.9.mhc_ffn.bias": "model.safetensors",
|
| 1018 |
+
"language_model.model.layers.9.mhc_ffn.weight": "model.safetensors",
|
| 1019 |
+
"language_model.model.layers.9.mlp.down_proj.biases": "model.safetensors",
|
| 1020 |
+
"language_model.model.layers.9.mlp.down_proj.scales": "model.safetensors",
|
| 1021 |
+
"language_model.model.layers.9.mlp.down_proj.weight": "model.safetensors",
|
| 1022 |
+
"language_model.model.layers.9.mlp.gate_proj.biases": "model.safetensors",
|
| 1023 |
+
"language_model.model.layers.9.mlp.gate_proj.scales": "model.safetensors",
|
| 1024 |
+
"language_model.model.layers.9.mlp.gate_proj.weight": "model.safetensors",
|
| 1025 |
+
"language_model.model.layers.9.mlp.up_proj.biases": "model.safetensors",
|
| 1026 |
+
"language_model.model.layers.9.mlp.up_proj.scales": "model.safetensors",
|
| 1027 |
+
"language_model.model.layers.9.mlp.up_proj.weight": "model.safetensors",
|
| 1028 |
+
"language_model.model.layers.9.post_attention_layernorm.weight": "model.safetensors",
|
| 1029 |
+
"language_model.model.norm.weight": "model.safetensors"
|
| 1030 |
+
}
|
| 1031 |
+
}
|
sensenova_loop.py
ADDED
|
@@ -0,0 +1,259 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Copyright © 2026 NetProjects (Paris Michalas). Apache-2.0.
|
| 2 |
+
# MLX port of SenseNova-Loop (sensenova/SenseNova-Loop-9B-Preview, modeling_sensenova_loop.py @ 9937e91).
|
| 3 |
+
# Not affiliated with SenseTime / SenseNova.
|
| 4 |
+
#
|
| 5 |
+
# SenseNova-Loop is Qwen3.5 (text) with two changes:
|
| 6 |
+
# 1. Looped execution: physical layers [start, end) run `eval_count` times with shared weights. Every execution
|
| 7 |
+
# ("visit") has its own cache slot (KV for full attention, conv + recurrent state for Gated DeltaNet).
|
| 8 |
+
# 2. Two-stream mHC (manifold-constrained hyper-connections) on layers [mhc.start, mhc.end): the hidden state is
|
| 9 |
+
# copied into `streams` residual streams at the first mHC layer and averaged back at the last one, on every pass.
|
| 10 |
+
# Each mHC layer has two fp32 routers (mhc_attn, mhc_ffn) that give pre / post gates and a Sinkhorn mixing matrix.
|
| 11 |
+
#
|
| 12 |
+
# Works as mlx_lm/models/sensenova_loop.py (model_type "sensenova_loop") or as a custom model_file with
|
| 13 |
+
# trust_remote_code=True.
|
| 14 |
+
|
| 15 |
+
import os
|
| 16 |
+
from dataclasses import dataclass, field
|
| 17 |
+
from functools import partial
|
| 18 |
+
from typing import Any, Dict, List, Optional, Tuple
|
| 19 |
+
|
| 20 |
+
import mlx.core as mx
|
| 21 |
+
import mlx.nn as nn
|
| 22 |
+
|
| 23 |
+
try:
|
| 24 |
+
from . import qwen3_5 as q35
|
| 25 |
+
from .base import BaseModelArgs, create_attention_mask, create_ssm_mask
|
| 26 |
+
from .cache import ArraysCache, KVCache
|
| 27 |
+
except ImportError: # loaded as a model_file
|
| 28 |
+
from mlx_lm.models import qwen3_5 as q35
|
| 29 |
+
from mlx_lm.models.base import BaseModelArgs, create_attention_mask, create_ssm_mask
|
| 30 |
+
from mlx_lm.models.cache import ArraysCache, KVCache
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
@dataclass
|
| 34 |
+
class ModelArgs(BaseModelArgs):
|
| 35 |
+
model_type: str
|
| 36 |
+
text_config: dict
|
| 37 |
+
loop_settings: dict = field(default_factory=dict)
|
| 38 |
+
mhc: dict = field(default_factory=dict)
|
| 39 |
+
|
| 40 |
+
|
| 41 |
+
def execution_plan(num_layers: int, loop: dict, mhc: dict) -> Tuple[int, ...]:
|
| 42 |
+
if not loop:
|
| 43 |
+
return tuple(range(num_layers))
|
| 44 |
+
start, end = int(loop["start"]), int(loop["end"])
|
| 45 |
+
count = int(loop.get("eval_count", (loop.get("counts") or [1])[0]))
|
| 46 |
+
if "counts" in loop and list(loop["counts"]) != [count]:
|
| 47 |
+
raise ValueError(f"loop_settings.counts {loop['counts']} must be [{count}]")
|
| 48 |
+
if not (0 <= start < end <= num_layers) or count < 1:
|
| 49 |
+
raise ValueError(f"bad loop_settings {loop}")
|
| 50 |
+
if mhc:
|
| 51 |
+
ms, me = int(mhc["start"]), int(mhc["end"])
|
| 52 |
+
inside = (start <= ms < me <= end) or (me <= start) or (ms >= end)
|
| 53 |
+
if not (0 <= ms < me <= num_layers) or not inside:
|
| 54 |
+
raise ValueError(f"mhc range [{ms}, {me}) must sit inside one loop phase")
|
| 55 |
+
return tuple(range(start)) + tuple(range(start, end)) * count + tuple(range(end, num_layers))
|
| 56 |
+
|
| 57 |
+
|
| 58 |
+
def sinkhorn(logits: mx.array, iterations: int) -> mx.array:
|
| 59 |
+
"""Log-space Sinkhorn: rows then columns, `iterations` times, then exp (fp32)."""
|
| 60 |
+
logits = logits.astype(mx.float32)
|
| 61 |
+
for _ in range(iterations):
|
| 62 |
+
logits = logits - mx.logsumexp(logits, axis=-1, keepdims=True)
|
| 63 |
+
logits = logits - mx.logsumexp(logits, axis=-2, keepdims=True)
|
| 64 |
+
return mx.exp(logits)
|
| 65 |
+
|
| 66 |
+
|
| 67 |
+
class MHCRouter(nn.Module):
|
| 68 |
+
"""fp32 router. Kept as raw parameters (not nn.Linear) so convert never quantizes it."""
|
| 69 |
+
|
| 70 |
+
def __init__(self, dim: int, streams: int, iterations: int, fixed_identity: bool, pre_scale: float):
|
| 71 |
+
super().__init__()
|
| 72 |
+
self.streams = streams
|
| 73 |
+
self.iterations = iterations
|
| 74 |
+
self.fixed_identity = fixed_identity
|
| 75 |
+
self.pre_scale = pre_scale
|
| 76 |
+
out = 2 * streams + (0 if fixed_identity else streams * streams)
|
| 77 |
+
self.weight = mx.zeros((out, streams * dim), dtype=mx.float32)
|
| 78 |
+
self.bias = mx.zeros((out,), dtype=mx.float32)
|
| 79 |
+
|
| 80 |
+
def __call__(self, residual: mx.array):
|
| 81 |
+
# residual: (..., streams, dim) in the model dtype
|
| 82 |
+
fn = _route_fn(self.streams, self.iterations, self.fixed_identity, self.pre_scale)
|
| 83 |
+
branch, post, mix = fn(residual, self.weight, self.bias)
|
| 84 |
+
return branch, post, (None if self.fixed_identity else mix)
|
| 85 |
+
|
| 86 |
+
@staticmethod
|
| 87 |
+
def merge(residual: mx.array, branch: mx.array, post: mx.array, mix: Optional[mx.array]) -> mx.array:
|
| 88 |
+
if mix is None:
|
| 89 |
+
return residual + post.astype(branch.dtype)[..., None] * branch[..., None, :]
|
| 90 |
+
return _merge_fn()(residual, branch, post, mix)
|
| 91 |
+
|
| 92 |
+
|
| 93 |
+
def _route(residual, weight, bias, s, iterations, fixed_identity, pre_scale):
|
| 94 |
+
flat = residual.reshape(*residual.shape[:-2], -1).astype(mx.float32)
|
| 95 |
+
normalized = flat * mx.rsqrt(mx.mean(mx.square(flat), axis=-1, keepdims=True) + 1e-6)
|
| 96 |
+
logits = normalized @ weight.astype(mx.float32).T + bias.astype(mx.float32)
|
| 97 |
+
pre = mx.sigmoid(logits[..., :s])
|
| 98 |
+
post = 2 * mx.sigmoid(logits[..., s : 2 * s])
|
| 99 |
+
if fixed_identity:
|
| 100 |
+
mix = mx.zeros((1,), dtype=mx.float32)
|
| 101 |
+
else:
|
| 102 |
+
mix = sinkhorn(logits[..., 2 * s :].reshape(*logits.shape[:-1], s, s), iterations)
|
| 103 |
+
branch = (pre.astype(residual.dtype)[..., None, :] @ residual).squeeze(-2)
|
| 104 |
+
if pre_scale != 1.0:
|
| 105 |
+
branch = branch * pre_scale
|
| 106 |
+
return branch, post, mix
|
| 107 |
+
|
| 108 |
+
|
| 109 |
+
def _merge(residual, branch, post, mix):
|
| 110 |
+
return mix.astype(residual.dtype) @ residual + post.astype(branch.dtype)[..., None] * branch[..., None, :]
|
| 111 |
+
|
| 112 |
+
|
| 113 |
+
# The router is ~45 tiny ops (20 Sinkhorn iterations) and runs 96 times per token, so it is compiled into one kernel
|
| 114 |
+
# graph; SN_NO_COMPILE=1 runs it eagerly (same math).
|
| 115 |
+
_COMPILE = os.environ.get("SN_NO_COMPILE", "0") != "1"
|
| 116 |
+
_ROUTE_CACHE: Dict[tuple, Any] = {}
|
| 117 |
+
|
| 118 |
+
|
| 119 |
+
def _route_fn(s, iterations, fixed_identity, pre_scale):
|
| 120 |
+
key = (s, iterations, fixed_identity, pre_scale)
|
| 121 |
+
if key not in _ROUTE_CACHE:
|
| 122 |
+
f = partial(_route, s=s, iterations=iterations, fixed_identity=fixed_identity, pre_scale=pre_scale)
|
| 123 |
+
_ROUTE_CACHE[key] = mx.compile(f) if _COMPILE else f
|
| 124 |
+
return _ROUTE_CACHE[key]
|
| 125 |
+
|
| 126 |
+
|
| 127 |
+
_MERGE = []
|
| 128 |
+
|
| 129 |
+
|
| 130 |
+
def _merge_fn():
|
| 131 |
+
if not _MERGE:
|
| 132 |
+
_MERGE.append(mx.compile(_merge) if _COMPILE else _merge)
|
| 133 |
+
return _MERGE[0]
|
| 134 |
+
|
| 135 |
+
|
| 136 |
+
class MHCDecoderLayer(q35.DecoderLayer):
|
| 137 |
+
def __init__(self, args: q35.TextModelArgs, layer_idx: int, mhc: dict):
|
| 138 |
+
super().__init__(args, layer_idx)
|
| 139 |
+
self.mhc_start, self.mhc_end = int(mhc["start"]), int(mhc["end"])
|
| 140 |
+
self.layer_idx = layer_idx
|
| 141 |
+
self.streams = int(mhc["streams"])
|
| 142 |
+
kw = dict(
|
| 143 |
+
dim=args.hidden_size,
|
| 144 |
+
streams=self.streams,
|
| 145 |
+
iterations=int(mhc["iterations"]),
|
| 146 |
+
fixed_identity=bool(mhc.get("fixed_identity", False)),
|
| 147 |
+
pre_scale=float(mhc.get("pre_scale", 1.0)),
|
| 148 |
+
)
|
| 149 |
+
self.mhc_attn = MHCRouter(**kw)
|
| 150 |
+
self.mhc_ffn = MHCRouter(**kw)
|
| 151 |
+
|
| 152 |
+
def __call__(self, x: mx.array, mask=None, cache=None) -> mx.array:
|
| 153 |
+
# enter: copy into streams at the first mHC layer, otherwise x already carries (B, L, streams, D)
|
| 154 |
+
if self.layer_idx == self.mhc_start:
|
| 155 |
+
residual = mx.broadcast_to(x[..., None, :], (*x.shape[:-1], self.streams, x.shape[-1]))
|
| 156 |
+
else:
|
| 157 |
+
residual = x
|
| 158 |
+
branch, post, mix = self.mhc_attn(residual)
|
| 159 |
+
branch = self.input_layernorm(branch)
|
| 160 |
+
if self.is_linear:
|
| 161 |
+
branch = self.linear_attn(branch, mask, cache)
|
| 162 |
+
else:
|
| 163 |
+
branch = self.self_attn(branch, mask, cache)
|
| 164 |
+
residual = MHCRouter.merge(residual, branch, post, mix)
|
| 165 |
+
branch, post, mix = self.mhc_ffn(residual)
|
| 166 |
+
branch = self.mlp(self.post_attention_layernorm(branch))
|
| 167 |
+
out = MHCRouter.merge(residual, branch, post, mix)
|
| 168 |
+
# leave: average the streams after the last mHC layer
|
| 169 |
+
return out.mean(axis=-2) if self.layer_idx == self.mhc_end - 1 else out
|
| 170 |
+
|
| 171 |
+
|
| 172 |
+
class SenseNovaLoopTextModel(nn.Module):
|
| 173 |
+
def __init__(self, args: q35.TextModelArgs, loop: dict, mhc: dict):
|
| 174 |
+
super().__init__()
|
| 175 |
+
self.args = args
|
| 176 |
+
self.embed_tokens = nn.Embedding(args.vocab_size, args.hidden_size)
|
| 177 |
+
ms, me = (int(mhc["start"]), int(mhc["end"])) if mhc else (0, 0)
|
| 178 |
+
self.layers = [
|
| 179 |
+
MHCDecoderLayer(args, i, mhc) if ms <= i < me else q35.DecoderLayer(args, i)
|
| 180 |
+
for i in range(args.num_hidden_layers)
|
| 181 |
+
]
|
| 182 |
+
self.norm = nn.RMSNorm(args.hidden_size, eps=args.rms_norm_eps)
|
| 183 |
+
self.plan = execution_plan(args.num_hidden_layers, loop, mhc)
|
| 184 |
+
self.fa_visit = next((e for e, p in enumerate(self.plan) if not self.layers[p].is_linear), None)
|
| 185 |
+
self.ssm_visit = next((e for e, p in enumerate(self.plan) if self.layers[p].is_linear), None)
|
| 186 |
+
|
| 187 |
+
def __call__(self, inputs: mx.array, cache=None, input_embeddings: Optional[mx.array] = None) -> mx.array:
|
| 188 |
+
h = input_embeddings if input_embeddings is not None else self.embed_tokens(inputs)
|
| 189 |
+
if cache is None:
|
| 190 |
+
cache = [None] * len(self.plan)
|
| 191 |
+
if len(cache) != len(self.plan):
|
| 192 |
+
raise ValueError(f"cache has {len(cache)} slots, the execution plan needs {len(self.plan)}")
|
| 193 |
+
fa_mask = create_attention_mask(h, cache[self.fa_visit]) if self.fa_visit is not None else None
|
| 194 |
+
ssm_mask = create_ssm_mask(h, cache[self.ssm_visit]) if self.ssm_visit is not None else None
|
| 195 |
+
for physical, c in zip(self.plan, cache):
|
| 196 |
+
layer = self.layers[physical]
|
| 197 |
+
h = layer(h, mask=ssm_mask if layer.is_linear else fa_mask, cache=c)
|
| 198 |
+
return self.norm(h)
|
| 199 |
+
|
| 200 |
+
|
| 201 |
+
class TextModel(q35.TextModel):
|
| 202 |
+
def __init__(self, args: q35.TextModelArgs, loop: dict, mhc: dict):
|
| 203 |
+
nn.Module.__init__(self)
|
| 204 |
+
self.args = args
|
| 205 |
+
self.model_type = args.model_type
|
| 206 |
+
self.model = SenseNovaLoopTextModel(args, loop, mhc)
|
| 207 |
+
if not args.tie_word_embeddings:
|
| 208 |
+
self.lm_head = nn.Linear(args.hidden_size, args.vocab_size, bias=False)
|
| 209 |
+
self.sharding_group = None
|
| 210 |
+
|
| 211 |
+
@property
|
| 212 |
+
def layers(self):
|
| 213 |
+
return self.model.layers
|
| 214 |
+
|
| 215 |
+
def make_cache(self):
|
| 216 |
+
return [ArraysCache(size=2) if self.model.layers[p].is_linear else KVCache() for p in self.model.plan]
|
| 217 |
+
|
| 218 |
+
@property
|
| 219 |
+
def cast_predicate(self):
|
| 220 |
+
def predicate(path: str):
|
| 221 |
+
return not (path.endswith("A_log") or ".mhc_attn." in path or ".mhc_ffn." in path)
|
| 222 |
+
|
| 223 |
+
return predicate
|
| 224 |
+
|
| 225 |
+
|
| 226 |
+
class Model(nn.Module):
|
| 227 |
+
def __init__(self, args: ModelArgs):
|
| 228 |
+
super().__init__()
|
| 229 |
+
self.args = args
|
| 230 |
+
self.model_type = args.model_type
|
| 231 |
+
self.language_model = TextModel(q35.TextModelArgs.from_dict(args.text_config), args.loop_settings, args.mhc)
|
| 232 |
+
|
| 233 |
+
def __call__(self, inputs: mx.array, cache=None, input_embeddings: Optional[mx.array] = None):
|
| 234 |
+
return self.language_model(inputs, cache=cache, input_embeddings=input_embeddings)
|
| 235 |
+
|
| 236 |
+
@property
|
| 237 |
+
def model(self):
|
| 238 |
+
return self.language_model.model
|
| 239 |
+
|
| 240 |
+
def sanitize(self, weights):
|
| 241 |
+
# q35: drops model.visual.* and mtp.*, renames model.language_model -> language_model.model, conv1d layout,
|
| 242 |
+
# +1 on norms for raw HF checkpoints. Routers stay fp32 (the reference refuses to run them in bf16).
|
| 243 |
+
weights = q35.Model.sanitize(self, weights)
|
| 244 |
+
return {k: (v.astype(mx.float32) if (".mhc_attn." in k or ".mhc_ffn." in k) else v) for k, v in weights.items()}
|
| 245 |
+
|
| 246 |
+
@property
|
| 247 |
+
def layers(self):
|
| 248 |
+
return self.language_model.model.layers
|
| 249 |
+
|
| 250 |
+
def make_cache(self):
|
| 251 |
+
return self.language_model.make_cache()
|
| 252 |
+
|
| 253 |
+
@property
|
| 254 |
+
def quant_predicate(self):
|
| 255 |
+
return None
|
| 256 |
+
|
| 257 |
+
@property
|
| 258 |
+
def cast_predicate(self):
|
| 259 |
+
return self.language_model.cast_predicate
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
|
| 3 |
+
size 19989325
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"local_files_only": false,
|
| 14 |
+
"model_max_length": 262144,
|
| 15 |
+
"model_specific_special_tokens": {
|
| 16 |
+
"audio_bos_token": "<|audio_start|>",
|
| 17 |
+
"audio_eos_token": "<|audio_end|>",
|
| 18 |
+
"audio_token": "<|audio_pad|>",
|
| 19 |
+
"image_token": "<|image_pad|>",
|
| 20 |
+
"video_token": "<|video_pad|>",
|
| 21 |
+
"vision_bos_token": "<|vision_start|>",
|
| 22 |
+
"vision_eos_token": "<|vision_end|>"
|
| 23 |
+
},
|
| 24 |
+
"pad_token": "<|endoftext|>",
|
| 25 |
+
"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+",
|
| 26 |
+
"processor_class": "Qwen3VLProcessor",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"tool_parser_type": "qwen3_coder",
|
| 30 |
+
"unk_token": null,
|
| 31 |
+
"video_token": "<|video_pad|>",
|
| 32 |
+
"vision_bos_token": "<|vision_start|>",
|
| 33 |
+
"vision_eos_token": "<|vision_end|>"
|
| 34 |
+
}
|