Add S1-mini by Superwhisper MLX 4-bit conversion
Browse files- .gitattributes +2 -0
- LICENSE +212 -0
- README.md +455 -0
- banner.jpg +3 -0
- chat_template.jinja +89 -0
- config.json +43 -0
- generation_config.json +10 -0
- model.safetensors +3 -0
- model.safetensors.index.json +712 -0
- tokenizer.json +3 -0
- tokenizer_config.json +31 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,5 @@ 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 |
+
banner.jpg filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
LICENSE
ADDED
|
@@ -0,0 +1,212 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Apache License
|
| 2 |
+
Version 2.0, January 2004
|
| 3 |
+
http://www.apache.org/licenses/
|
| 4 |
+
|
| 5 |
+
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
|
| 6 |
+
|
| 7 |
+
1. Definitions.
|
| 8 |
+
|
| 9 |
+
"License" shall mean the terms and conditions for use, reproduction,
|
| 10 |
+
and distribution as defined by Sections 1 through 9 of this document.
|
| 11 |
+
|
| 12 |
+
"Licensor" shall mean the copyright owner or entity authorized by
|
| 13 |
+
the copyright owner that is granting the License.
|
| 14 |
+
|
| 15 |
+
"Legal Entity" shall mean the union of the acting entity and all
|
| 16 |
+
other entities that control, are controlled by, or are under common
|
| 17 |
+
control with that entity. For the purposes of this definition,
|
| 18 |
+
"control" means (i) the power, direct or indirect, to cause the
|
| 19 |
+
direction or management of such entity, whether by contract or
|
| 20 |
+
otherwise, or (ii) ownership of fifty percent (50%) or more of the
|
| 21 |
+
outstanding shares, or (iii) beneficial ownership of such entity.
|
| 22 |
+
|
| 23 |
+
"You" (or "Your") shall mean an individual or Legal Entity
|
| 24 |
+
exercising permissions granted by this License.
|
| 25 |
+
|
| 26 |
+
"Source" form shall mean the preferred form for making modifications,
|
| 27 |
+
including but not limited to software source code, documentation
|
| 28 |
+
source, and configuration files.
|
| 29 |
+
|
| 30 |
+
"Object" form shall mean any form resulting from mechanical
|
| 31 |
+
transformation or translation of a Source form, including but
|
| 32 |
+
not limited to compiled object code, generated documentation,
|
| 33 |
+
and conversions to other media types.
|
| 34 |
+
|
| 35 |
+
"Work" shall mean the work of authorship, whether in Source or
|
| 36 |
+
Object form, made available under the License, as indicated by a
|
| 37 |
+
copyright notice that is included in or attached to the work
|
| 38 |
+
(an example is provided in the Appendix below).
|
| 39 |
+
|
| 40 |
+
"Derivative Works" shall mean any work, whether in Source or Object
|
| 41 |
+
form, that is based on (or derived from) the Work and for which the
|
| 42 |
+
editorial revisions, annotations, elaborations, or other modifications
|
| 43 |
+
represent, as a whole, an original work of authorship. For the purposes
|
| 44 |
+
of this License, Derivative Works shall not include works that remain
|
| 45 |
+
separable from, or merely link (or bind by name) to the interfaces of,
|
| 46 |
+
the Work and Derivative Works thereof.
|
| 47 |
+
|
| 48 |
+
"Contribution" shall mean any work of authorship, including
|
| 49 |
+
the original version of the Work and any modifications or additions
|
| 50 |
+
to that Work or Derivative Works thereof, that is intentionally
|
| 51 |
+
submitted to Licensor for inclusion in the Work by the copyright owner
|
| 52 |
+
or by an individual or Legal Entity authorized to submit on behalf of
|
| 53 |
+
the copyright owner. For the purposes of this definition, "submitted"
|
| 54 |
+
means any form of electronic, verbal, or written communication sent
|
| 55 |
+
to the Licensor or its representatives, including but not limited to
|
| 56 |
+
communication on electronic mailing lists, source code control systems,
|
| 57 |
+
and issue tracking systems that are managed by, or on behalf of, the
|
| 58 |
+
Licensor for the purpose of discussing and improving the Work, but
|
| 59 |
+
excluding communication that is conspicuously marked or otherwise
|
| 60 |
+
designated in writing by the copyright owner as "Not a Contribution."
|
| 61 |
+
|
| 62 |
+
"Contributor" shall mean Licensor and any individual or Legal Entity
|
| 63 |
+
on behalf of whom a Contribution has been received by Licensor and
|
| 64 |
+
subsequently incorporated within the Work.
|
| 65 |
+
|
| 66 |
+
2. Grant of Copyright License. Subject to the terms and conditions of
|
| 67 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 68 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 69 |
+
copyright license to reproduce, prepare Derivative Works of,
|
| 70 |
+
publicly display, publicly perform, sublicense, and distribute the
|
| 71 |
+
Work and such Derivative Works in Source or Object form.
|
| 72 |
+
|
| 73 |
+
3. Grant of Patent License. Subject to the terms and conditions of
|
| 74 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 75 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 76 |
+
(except as stated in this section) patent license to make, have made,
|
| 77 |
+
use, offer to sell, sell, import, and otherwise transfer the Work,
|
| 78 |
+
where such license applies only to those patent claims licensable
|
| 79 |
+
by such Contributor that are necessarily infringed by their
|
| 80 |
+
Contribution(s) alone or by combination of their Contribution(s)
|
| 81 |
+
with the Work to which such Contribution(s) was submitted. If You
|
| 82 |
+
institute patent litigation against any entity (including a
|
| 83 |
+
cross-claim or counterclaim in a lawsuit) alleging that the Work
|
| 84 |
+
or a Contribution incorporated within the Work constitutes direct
|
| 85 |
+
or contributory patent infringement, then any patent licenses
|
| 86 |
+
granted to You under this License for that Work shall terminate
|
| 87 |
+
as of the date such litigation is filed.
|
| 88 |
+
|
| 89 |
+
4. Redistribution. You may reproduce and distribute copies of the
|
| 90 |
+
Work or Derivative Works thereof in any medium, with or without
|
| 91 |
+
modifications, and in Source or Object form, provided that You
|
| 92 |
+
meet the following conditions:
|
| 93 |
+
|
| 94 |
+
(a) You must give any other recipients of the Work or
|
| 95 |
+
Derivative Works a copy of this License; and
|
| 96 |
+
|
| 97 |
+
(b) You must cause any modified files to carry prominent notices
|
| 98 |
+
stating that You changed the files; and
|
| 99 |
+
|
| 100 |
+
(c) You must retain, in the Source form of any Derivative Works
|
| 101 |
+
that You distribute, all copyright, patent, trademark, and
|
| 102 |
+
attribution notices from the Source form of the Work,
|
| 103 |
+
excluding those notices that do not pertain to any part of
|
| 104 |
+
the Derivative Works; and
|
| 105 |
+
|
| 106 |
+
(d) If the Work includes a "NOTICE" text file as part of its
|
| 107 |
+
distribution, then any Derivative Works that You distribute must
|
| 108 |
+
include a readable copy of the attribution notices contained
|
| 109 |
+
within such NOTICE file, excluding those notices that do not
|
| 110 |
+
pertain to any part of the Derivative Works, in at least one
|
| 111 |
+
of the following places: within a NOTICE text file distributed
|
| 112 |
+
as part of the Derivative Works; within the Source form or
|
| 113 |
+
documentation, if provided along with the Derivative Works; or,
|
| 114 |
+
within a display generated by the Derivative Works, if and
|
| 115 |
+
wherever such third-party notices normally appear. The contents
|
| 116 |
+
of the NOTICE file are for informational purposes only and
|
| 117 |
+
do not modify the License. You may add Your own attribution
|
| 118 |
+
notices within Derivative Works that You distribute, alongside
|
| 119 |
+
or as an addendum to the NOTICE text from the Work, provided
|
| 120 |
+
that such additional attribution notices cannot be construed
|
| 121 |
+
as modifying the License.
|
| 122 |
+
|
| 123 |
+
You may add Your own copyright statement to Your modifications and
|
| 124 |
+
may provide additional or different license terms and conditions
|
| 125 |
+
for use, reproduction, or distribution of Your modifications, or
|
| 126 |
+
for any such Derivative Works as a whole, provided Your use,
|
| 127 |
+
reproduction, and distribution of the Work otherwise complies with
|
| 128 |
+
the conditions stated in this License.
|
| 129 |
+
|
| 130 |
+
5. Submission of Contributions. Unless You explicitly state otherwise,
|
| 131 |
+
any Contribution intentionally submitted for inclusion in the Work
|
| 132 |
+
by You to the Licensor shall be under the terms and conditions of
|
| 133 |
+
this License, without any additional terms or conditions.
|
| 134 |
+
Notwithstanding the above, nothing herein shall supersede or modify
|
| 135 |
+
the terms of any separate license agreement you may have executed
|
| 136 |
+
with Licensor regarding such Contributions.
|
| 137 |
+
|
| 138 |
+
6. Trademarks. This License does not grant permission to use the trade
|
| 139 |
+
names, trademarks, service marks, or product names of the Licensor,
|
| 140 |
+
except as required for reasonable and customary use in describing the
|
| 141 |
+
origin of the Work and reproducing the content of the NOTICE file.
|
| 142 |
+
|
| 143 |
+
7. Disclaimer of Warranty. Unless required by applicable law or
|
| 144 |
+
agreed to in writing, Licensor provides the Work (and each
|
| 145 |
+
Contributor provides its Contributions) on an "AS IS" BASIS,
|
| 146 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
| 147 |
+
implied, including, without limitation, any warranties or conditions
|
| 148 |
+
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
|
| 149 |
+
PARTICULAR PURPOSE. You are solely responsible for determining the
|
| 150 |
+
appropriateness of using or redistributing the Work and assume any
|
| 151 |
+
risks associated with Your exercise of permissions under this License.
|
| 152 |
+
|
| 153 |
+
8. Limitation of Liability. In no event and under no legal theory,
|
| 154 |
+
whether in tort (including negligence), contract, or otherwise,
|
| 155 |
+
unless required by applicable law (such as deliberate and grossly
|
| 156 |
+
negligent acts) or agreed to in writing, shall any Contributor be
|
| 157 |
+
liable to You for damages, including any direct, indirect, special,
|
| 158 |
+
incidental, or consequential damages of any character arising as a
|
| 159 |
+
result of this License or out of the use or inability to use the
|
| 160 |
+
Work (including but not limited to damages for loss of goodwill,
|
| 161 |
+
work stoppage, computer failure or malfunction, or any and all
|
| 162 |
+
other commercial damages or losses), even if such Contributor
|
| 163 |
+
has been advised of the possibility of such damages.
|
| 164 |
+
|
| 165 |
+
9. Accepting Warranty or Additional Liability. While redistributing
|
| 166 |
+
the Work or Derivative Works thereof, You may choose to offer,
|
| 167 |
+
and charge a fee for, acceptance of support, warranty, indemnity,
|
| 168 |
+
or other liability obligations and/or rights consistent with this
|
| 169 |
+
License. However, in accepting such obligations, You may act only
|
| 170 |
+
on Your own behalf and on Your sole responsibility, not on behalf
|
| 171 |
+
of any other Contributor, and only if You agree to indemnify,
|
| 172 |
+
defend, and hold each Contributor harmless for any liability
|
| 173 |
+
incurred by, or claims asserted against, such Contributor by reason
|
| 174 |
+
of your accepting any such warranty or additional liability.
|
| 175 |
+
|
| 176 |
+
END OF TERMS AND CONDITIONS
|
| 177 |
+
|
| 178 |
+
APPENDIX: How to apply the Apache License to your work.
|
| 179 |
+
|
| 180 |
+
To apply the Apache License to your work, attach the following
|
| 181 |
+
boilerplate notice, with the fields enclosed by brackets "[]"
|
| 182 |
+
replaced with your own identifying information. (Don't include
|
| 183 |
+
the brackets!) The text should be enclosed in the appropriate
|
| 184 |
+
comment syntax for the file format. We also recommend that a
|
| 185 |
+
file or class name and description of purpose be included on the
|
| 186 |
+
same "printed page" as the copyright notice for easier
|
| 187 |
+
identification within third-party archives.
|
| 188 |
+
|
| 189 |
+
Copyright [yyyy] [name of copyright owner]
|
| 190 |
+
|
| 191 |
+
Licensed under the Apache License, Version 2.0 (the "License");
|
| 192 |
+
you may not use this file except in compliance with the License.
|
| 193 |
+
You may obtain a copy of the License at
|
| 194 |
+
|
| 195 |
+
http://www.apache.org/licenses/LICENSE-2.0
|
| 196 |
+
|
| 197 |
+
Unless required by applicable law or agreed to in writing, software
|
| 198 |
+
distributed under the License is distributed on an "AS IS" BASIS,
|
| 199 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
| 200 |
+
See the License for the specific language governing permissions and
|
| 201 |
+
limitations under the License.
|
| 202 |
+
|
| 203 |
+
------------------------------------------------------------------------
|
| 204 |
+
|
| 205 |
+
ADDITIONAL TERM
|
| 206 |
+
|
| 207 |
+
In addition to the terms of the Apache License, Version 2.0 above: any
|
| 208 |
+
use, distribution, or integration of this model, whether unmodified or
|
| 209 |
+
as part of a derivative work or product, must continue to identify it
|
| 210 |
+
by its original name, "S1-mini" by "Superwhisper", using that exact
|
| 211 |
+
capitalization, regardless of any other name under which the model or
|
| 212 |
+
a product incorporating it is marketed or distributed.
|
README.md
ADDED
|
@@ -0,0 +1,455 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
license_name: s1-mini-license
|
| 4 |
+
license_link: LICENSE
|
| 5 |
+
base_model: superwhisper/s1-mini
|
| 6 |
+
base_model_relation: quantized
|
| 7 |
+
library_name: mlx
|
| 8 |
+
pipeline_tag: text-generation
|
| 9 |
+
language:
|
| 10 |
+
- en
|
| 11 |
+
tags:
|
| 12 |
+
- asr
|
| 13 |
+
- text-normalization
|
| 14 |
+
- inverse-text-normalization
|
| 15 |
+
- punctuation
|
| 16 |
+
- truecasing
|
| 17 |
+
- speech-to-text
|
| 18 |
+
- dictation
|
| 19 |
+
- post-processing
|
| 20 |
+
- qwen3
|
| 21 |
+
- mlx
|
| 22 |
+
- 4-bit
|
| 23 |
+
---
|
| 24 |
+
|
| 25 |
+
# S1-mini by [Superwhisper](https://superwhisper.com) — MLX 4-bit
|
| 26 |
+
|
| 27 |
+
<div align="center">
|
| 28 |
+
<img src="./banner.jpg" alt="S1-mini banner" width="100%">
|
| 29 |
+
|
| 30 |
+
[](https://superwhisper.com)
|
| 31 |
+
[](https://discord.gg/tF98XvJNvB)
|
| 32 |
+
[](https://huggingface.co/superwhisper/s1-mini/tree/v1)
|
| 33 |
+
</div>
|
| 34 |
+
|
| 35 |
+
> This repository contains an MLX affine 4-bit conversion of
|
| 36 |
+
> [S1-mini by Superwhisper](https://huggingface.co/superwhisper/s1-mini),
|
| 37 |
+
> converted from the original BF16 weights with `mlx-lm`. It uses a group size
|
| 38 |
+
> of 64 and preserves the original tokenizer, chat template, generation
|
| 39 |
+
> configuration, model documentation, and license.
|
| 40 |
+
|
| 41 |
+
## Use with MLX
|
| 42 |
+
|
| 43 |
+
```bash
|
| 44 |
+
uv tool install mlx-lm
|
| 45 |
+
```
|
| 46 |
+
|
| 47 |
+
```python
|
| 48 |
+
from mlx_lm import generate, load
|
| 49 |
+
|
| 50 |
+
model, tokenizer = load("mlx-community/S1-mini-MLX-4bit")
|
| 51 |
+
system = (
|
| 52 |
+
"You are a text normalizer for speech-to-text transcripts. The input begins "
|
| 53 |
+
"with a control line specifying the styling, structure, and context settings; "
|
| 54 |
+
"clean the transcript to match those settings and output only the cleaned text."
|
| 55 |
+
)
|
| 56 |
+
messages = [
|
| 57 |
+
{"role": "system", "content": system},
|
| 58 |
+
{"role": "user", "content": "[Styling: semi-formal] [Structure: prose] [Context: general]\n"
|
| 59 |
+
"so um send the report by uh friday"},
|
| 60 |
+
]
|
| 61 |
+
prompt = tokenizer.apply_chat_template(
|
| 62 |
+
messages, tokenize=False, add_generation_prompt=True, enable_thinking=False
|
| 63 |
+
)
|
| 64 |
+
print(generate(model, tokenizer, prompt=prompt, max_tokens=128, verbose=False))
|
| 65 |
+
```
|
| 66 |
+
|
| 67 |
+
Converted with `mlx-lm` 0.31.3. A six-case deterministic sanity check matched
|
| 68 |
+
three upstream reference strings exactly. The remaining differences included a
|
| 69 |
+
retained leading “So,” punctuation/ordinal variation, and omission of “tomorrow”
|
| 70 |
+
in one correction case. This is a small functional check, not the upstream
|
| 71 |
+
7,519-case evaluation; assess the 4-bit build on your own transcripts.
|
| 72 |
+
|
| 73 |
+
A 0.6B-parameter text normalizer for speech-to-text output. It takes a raw ASR
|
| 74 |
+
transcript and rewrites it as clean written text: fillers removed, false starts
|
| 75 |
+
and self-corrections resolved to the value the speaker landed on, punctuation
|
| 76 |
+
and capitalization applied, and spoken numbers, dates, times, currency and
|
| 77 |
+
email addresses rendered in written form.
|
| 78 |
+
|
| 79 |
+
On a held-out set of 7,519 English cases it reaches 94.8% token accuracy, and
|
| 80 |
+
the quantized build is a 462 MiB file that runs comfortably on a laptop CPU.
|
| 81 |
+
|
| 82 |
+
This is release v1, and it covers English only. S1-mini is not a chat model
|
| 83 |
+
and will not follow general instructions; it does one job, and you steer it
|
| 84 |
+
with a control line at the top of the input.
|
| 85 |
+
|
| 86 |
+
Fine-tuned from [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B). If
|
| 87 |
+
you want to run it in llama.cpp, Ollama, LM Studio, or anything else built on
|
| 88 |
+
llama.cpp, grab the GGUF builds from
|
| 89 |
+
[superwhisper/s1-mini-GGUF](https://huggingface.co/superwhisper/s1-mini-GGUF).
|
| 90 |
+
You can use it in your own dictation app too, just check the license first.
|
| 91 |
+
|
| 92 |
+
Releases are tagged, so you can pin one:
|
| 93 |
+
`from_pretrained("superwhisper/s1-mini", revision="v1")`.
|
| 94 |
+
|
| 95 |
+
## Model overview
|
| 96 |
+
|
| 97 |
+
| | |
|
| 98 |
+
|---|---|
|
| 99 |
+
| Type | Causal language model, fine-tuned for a single transformation task |
|
| 100 |
+
| Base model | [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B) |
|
| 101 |
+
| Parameters | 596M total (0.44B non-embedding), embeddings tied |
|
| 102 |
+
| Layers | 28 |
|
| 103 |
+
| Attention heads | 16 for Q, 8 for KV (GQA) |
|
| 104 |
+
| Precision | BF16 |
|
| 105 |
+
| Recommended input | Up to ~1,000 tokens; chunk longer transcripts |
|
| 106 |
+
| Language | English |
|
| 107 |
+
| License | Apache 2.0 + naming clause ([LICENSE](LICENSE)) |
|
| 108 |
+
|
| 109 |
+
> [!NOTE]
|
| 110 |
+
> The Hub sidebar reports 0.8B parameters for this repo. `config.json` sets
|
| 111 |
+
> `tie_word_embeddings`, but `model.safetensors` still stores `lm_head.weight`
|
| 112 |
+
> as a materialized copy of the input embedding, so the 155.6M-parameter
|
| 113 |
+
> embedding is counted twice: 751.6M tensor elements against 596.0M unique
|
| 114 |
+
> parameters. The layout is inherited from `Qwen/Qwen3-0.6B`, which reports 0.8B
|
| 115 |
+
> on the Hub for the same reason. The table above counts unique parameters.
|
| 116 |
+
|
| 117 |
+
**Input.** The model expects the system prompt, then a control line, a
|
| 118 |
+
newline, and one raw ASR transcript, which will usually arrive lowercase and
|
| 119 |
+
unpunctuated. That is the shape it was trained on.
|
| 120 |
+
|
| 121 |
+
**Output.** It returns the cleaned transcript as plain text and nothing else,
|
| 122 |
+
with no preamble and no explanation. Under `Structure: lists` the output may
|
| 123 |
+
contain Markdown bullets, and under `Context: email` it may contain blank
|
| 124 |
+
lines separating a greeting, body and sign-off. When the input is nothing but
|
| 125 |
+
filler or noise, the correct output is an empty string, and that is what you
|
| 126 |
+
get.
|
| 127 |
+
|
| 128 |
+
## Source-model Transformers quickstart
|
| 129 |
+
|
| 130 |
+
Qwen3 support landed in `transformers` 4.51.0; with anything older you will
|
| 131 |
+
get `KeyError: 'qwen3'`.
|
| 132 |
+
|
| 133 |
+
```bash
|
| 134 |
+
pip install "transformers>=4.51.0" torch
|
| 135 |
+
```
|
| 136 |
+
|
| 137 |
+
```python
|
| 138 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 139 |
+
|
| 140 |
+
MODEL = "superwhisper/s1-mini"
|
| 141 |
+
|
| 142 |
+
tok = AutoTokenizer.from_pretrained(MODEL)
|
| 143 |
+
model = AutoModelForCausalLM.from_pretrained(MODEL, torch_dtype="auto")
|
| 144 |
+
|
| 145 |
+
# Required. Use this exact system prompt.
|
| 146 |
+
SYSTEM = (
|
| 147 |
+
"You are a text normalizer for speech-to-text transcripts. The input begins "
|
| 148 |
+
"with a control line specifying the styling, structure, and context settings; "
|
| 149 |
+
"clean the transcript to match those settings and output only the cleaned text."
|
| 150 |
+
)
|
| 151 |
+
|
| 152 |
+
def normalize(transcript, styling="semi-formal", structure="prose", context="general"):
|
| 153 |
+
control = f"[Styling: {styling}] [Structure: {structure}] [Context: {context}]"
|
| 154 |
+
messages = [
|
| 155 |
+
{"role": "system", "content": SYSTEM},
|
| 156 |
+
{"role": "user", "content": f"{control}\n{transcript}"},
|
| 157 |
+
]
|
| 158 |
+
text = tok.apply_chat_template(
|
| 159 |
+
messages,
|
| 160 |
+
tokenize=False,
|
| 161 |
+
add_generation_prompt=True,
|
| 162 |
+
enable_thinking=False, # required, see below
|
| 163 |
+
)
|
| 164 |
+
inputs = tok(text, return_tensors="pt").to(model.device)
|
| 165 |
+
out = model.generate(**inputs, max_new_tokens=1024, do_sample=False)
|
| 166 |
+
return tok.decode(out[0][inputs.input_ids.shape[1]:], skip_special_tokens=True)
|
| 167 |
+
|
| 168 |
+
|
| 169 |
+
raw = "so um i need to like send the the report by uh friday no wait make that thursday"
|
| 170 |
+
print(normalize(raw))
|
| 171 |
+
# I need to send the report by Thursday.
|
| 172 |
+
```
|
| 173 |
+
|
| 174 |
+
`torch_dtype="auto"` picks up the BF16 in `config.json`. At 0.6B this runs
|
| 175 |
+
comfortably on CPU; add `device_map="auto"` to place it on a GPU automatically
|
| 176 |
+
(requires `accelerate`).
|
| 177 |
+
|
| 178 |
+
## The control line
|
| 179 |
+
|
| 180 |
+
Every input starts with a control line, then a newline, then the transcript:
|
| 181 |
+
|
| 182 |
+
```
|
| 183 |
+
[Styling: <value>] [Structure: <value>] [Context: <value>]
|
| 184 |
+
<raw transcript>
|
| 185 |
+
```
|
| 186 |
+
|
| 187 |
+
| Axis | Values | What it does |
|
| 188 |
+
|---|---|---|
|
| 189 |
+
| `Styling` | `casual`, `semi-casual`, `semi-formal`, `formal` | Sets the register: how much capitalization, apostrophe and contraction cleanup to apply. |
|
| 190 |
+
| `Structure` | `prose`, `lists` | Whether the model may break enumerable content into a bulleted list. It needs at least three items, and anything that isn't really a list stays as prose. |
|
| 191 |
+
| `Context` | `general`, `email` | Destination conventions. `email` turns on greeting-line and sign-off-block layout. |
|
| 192 |
+
|
| 193 |
+
The three axes are independent and every combination was trained.
|
| 194 |
+
|
| 195 |
+
> [!IMPORTANT]
|
| 196 |
+
> The system prompt and the control line are part of the input format the
|
| 197 |
+
> model was trained on. Skip either one, change the system prompt's wording,
|
| 198 |
+
> or send values outside the trained sets, and the model can hallucinate or
|
| 199 |
+
> produce garbled output. Always send both, exactly as shown.
|
| 200 |
+
|
| 201 |
+
### Styling
|
| 202 |
+
|
| 203 |
+
The register decides how much of the speaker's voice survives into the
|
| 204 |
+
written text.
|
| 205 |
+
|
| 206 |
+
| Value | Behavior |
|
| 207 |
+
|---|---|
|
| 208 |
+
| `casual` | Everything lowercase, apostrophes stripped, colloquialisms kept, final period usually omitted. |
|
| 209 |
+
| `semi-casual` | Keeps the speaker's phrasing. `I` and its contractions are capitalized, sentence starts stay lowercase, final period usually omitted. |
|
| 210 |
+
| `semi-formal` | Standard written English: full capitalization and punctuation, contractions kept, colloquialisms smoothed (`gonna` becomes `going to`). A good default. |
|
| 211 |
+
| `formal` | Like `semi-formal`, with contractions expanded (`I am`, `cannot`). |
|
| 212 |
+
|
| 213 |
+
Here is the same input under all four registers:
|
| 214 |
+
|
| 215 |
+
Input: `hmm im gonna be late theres a cute dog outside i cant just walk past him`
|
| 216 |
+
|
| 217 |
+
| Styling | Output |
|
| 218 |
+
|---|---|
|
| 219 |
+
| `casual` | `hmm im gonna be late. theres a cute dog outside. i cant just walk past him` |
|
| 220 |
+
| `semi-casual` | `hmm, I'm gonna be late. there's a cute dog outside. I can't just walk past him` |
|
| 221 |
+
| `semi-formal` | `I'm going to be late. There's a cute dog outside. I can't just walk past him.` |
|
| 222 |
+
| `formal` | `I am going to be late. There is a cute dog outside. I cannot just walk past him.` |
|
| 223 |
+
|
| 224 |
+
Filled pauses like `um` and `uh` are removed in every register. Apostrophes
|
| 225 |
+
are decided by the register rather than the raw transcript, so the input can
|
| 226 |
+
arrive as `im` or `I'm` and the output comes out the same either way.
|
| 227 |
+
|
| 228 |
+
### Structure
|
| 229 |
+
|
| 230 |
+
`prose` keeps everything in sentences and paragraphs. `lists` permits the
|
| 231 |
+
model to break enumerable content into Markdown bullets, and it is
|
| 232 |
+
deliberately conservative about it: it wants at least three items, and
|
| 233 |
+
content that is not clearly an enumeration stays as prose. Here is the same
|
| 234 |
+
input under both values:
|
| 235 |
+
|
| 236 |
+
Input: `so for the trip we need to pack sunscreen and then also a first aid kit and um chargers for everything`
|
| 237 |
+
|
| 238 |
+
`Structure: prose`
|
| 239 |
+
|
| 240 |
+
```
|
| 241 |
+
So for the trip, we need to pack sunscreen and then also a first aid kit and chargers for everything.
|
| 242 |
+
```
|
| 243 |
+
|
| 244 |
+
`Structure: lists`
|
| 245 |
+
|
| 246 |
+
```
|
| 247 |
+
So for the trip, we need to pack:
|
| 248 |
+
- Sunscreen
|
| 249 |
+
- A first aid kit
|
| 250 |
+
- Chargers for everything
|
| 251 |
+
```
|
| 252 |
+
|
| 253 |
+
### Context
|
| 254 |
+
|
| 255 |
+
`general` produces flowing text, while `email` reshapes the transcript into
|
| 256 |
+
email layout, with a greeting line, the body and a sign-off block separated by
|
| 257 |
+
blank lines. Here is the same input under both values:
|
| 258 |
+
|
| 259 |
+
Input: `hey sarah just wanted to follow up on the proposal can you send the numbers by end of week thanks john`
|
| 260 |
+
|
| 261 |
+
`Context: general`
|
| 262 |
+
|
| 263 |
+
```
|
| 264 |
+
Hey Sarah, just wanted to follow up on the proposal. Can you send the numbers by end of week? Thanks, John.
|
| 265 |
+
```
|
| 266 |
+
|
| 267 |
+
`Context: email`
|
| 268 |
+
|
| 269 |
+
```
|
| 270 |
+
Hey Sarah,
|
| 271 |
+
|
| 272 |
+
Just wanted to follow up on the proposal. Can you send the numbers by end of week?
|
| 273 |
+
|
| 274 |
+
Thanks,
|
| 275 |
+
John
|
| 276 |
+
```
|
| 277 |
+
|
| 278 |
+
## Examples
|
| 279 |
+
|
| 280 |
+
All of these were measured on the BF16 weights in this repo with greedy
|
| 281 |
+
decoding, under `[Styling: semi-formal] [Structure: prose] [Context: general]`.
|
| 282 |
+
|
| 283 |
+
| Input | Output |
|
| 284 |
+
|---|---|
|
| 285 |
+
| `so um i need to like send the the report by uh friday no wait make that thursday` | `I need to send the report by Thursday.` |
|
| 286 |
+
| `i think the answer is forty two no sorry forty three` | `I think the answer is 43.` |
|
| 287 |
+
| `let's meet at half past two tomorrow uh actually make it three fifteen p m` | `Let's meet at 3:15pm tomorrow.` |
|
| 288 |
+
| `the invoice came to twenty three thousand four hundred and fifty dollars and it's due on march third twenty twenty six` | `The invoice came to $23,450, and it's due on March 3, 2026.` |
|
| 289 |
+
| `send it to support at superwhisper dot com` | `Send it to support@superwhisper.com.` |
|
| 290 |
+
| `um` | *(empty string)* |
|
| 291 |
+
|
| 292 |
+
## Set `enable_thinking=False`
|
| 293 |
+
|
| 294 |
+
The chat template comes from Qwen3 unchanged, and Qwen3 turns on thinking mode
|
| 295 |
+
by default. S1-mini was trained with thinking off and has no reasoning traces
|
| 296 |
+
in its training data.
|
| 297 |
+
|
| 298 |
+
> [!WARNING]
|
| 299 |
+
> If you leave the flag out you will usually get **no usable output at all**:
|
| 300 |
+
> the model emits an empty `<think>` block and stops. This is the single most
|
| 301 |
+
> common way to get a blank result from this model.
|
| 302 |
+
|
| 303 |
+
The flag makes the template emit an empty think block before the assistant
|
| 304 |
+
turn. That is the exact prefix the model saw during training. The template
|
| 305 |
+
ships as `chat_template.jinja` in this repo, so `apply_chat_template` picks it
|
| 306 |
+
up with no extra configuration. If you build prompts by hand instead, the
|
| 307 |
+
literal string is:
|
| 308 |
+
|
| 309 |
+
```
|
| 310 |
+
<|im_start|>system
|
| 311 |
+
You are a text normalizer for speech-to-text transcripts. The input begins with a control line specifying the styling, structure, and context settings; clean the transcript to match those settings and output only the cleaned text.<|im_end|>
|
| 312 |
+
<|im_start|>user
|
| 313 |
+
[Styling: semi-formal] [Structure: prose] [Context: general]
|
| 314 |
+
<raw transcript><|im_end|>
|
| 315 |
+
<|im_start|>assistant
|
| 316 |
+
<think>
|
| 317 |
+
|
| 318 |
+
</think>
|
| 319 |
+
|
| 320 |
+
```
|
| 321 |
+
|
| 322 |
+
Written out, the assistant prefix is
|
| 323 |
+
`<|im_start|>assistant\n<think>\n\n</think>\n\n`, with two newlines inside the
|
| 324 |
+
think block and two more after it.
|
| 325 |
+
|
| 326 |
+
## Deployment
|
| 327 |
+
|
| 328 |
+
**vLLM.** Serve normally, then disable thinking per request:
|
| 329 |
+
|
| 330 |
+
```bash
|
| 331 |
+
vllm serve superwhisper/s1-mini
|
| 332 |
+
```
|
| 333 |
+
|
| 334 |
+
```bash
|
| 335 |
+
curl http://localhost:8000/v1/chat/completions -H "Content-Type: application/json" -d '{
|
| 336 |
+
"model": "superwhisper/s1-mini",
|
| 337 |
+
"messages": [
|
| 338 |
+
{"role": "system", "content": "You are a text normalizer for speech-to-text transcripts. The input begins with a control line specifying the styling, structure, and context settings; clean the transcript to match those settings and output only the cleaned text."},
|
| 339 |
+
{"role": "user", "content": "[Styling: semi-formal] [Structure: prose] [Context: general]\nso um send the report by uh friday"}
|
| 340 |
+
],
|
| 341 |
+
"temperature": 0,
|
| 342 |
+
"chat_template_kwargs": {"enable_thinking": false}
|
| 343 |
+
}'
|
| 344 |
+
```
|
| 345 |
+
|
| 346 |
+
**SGLang.** The same request shape works here, since `chat_template_kwargs`
|
| 347 |
+
is supported on the OpenAI-compatible endpoint.
|
| 348 |
+
|
| 349 |
+
```bash
|
| 350 |
+
python -m sglang.launch_server --model-path superwhisper/s1-mini
|
| 351 |
+
```
|
| 352 |
+
|
| 353 |
+
**llama.cpp.** Use the [GGUF
|
| 354 |
+
builds](https://huggingface.co/superwhisper/s1-mini-GGUF). The template must be
|
| 355 |
+
applied with thinking disabled, which means passing `--jinja` together with
|
| 356 |
+
`--chat-template-kwargs`. Don't substitute `--reasoning-budget 0`, which
|
| 357 |
+
suppresses the think block a different way and degrades the output:
|
| 358 |
+
|
| 359 |
+
```bash
|
| 360 |
+
llama-server -hf superwhisper/s1-mini-GGUF:Q4_K_M --jinja --chat-template-kwargs '{"enable_thinking":false}' --temp 0
|
| 361 |
+
```
|
| 362 |
+
|
| 363 |
+
**Ollama and LM Studio.** Both work from the GGUF builds, but both inherit
|
| 364 |
+
Qwen3's thinking-on-by-default template. Make sure the assistant turn begins
|
| 365 |
+
with the empty think block shown above, or you will get blank output.
|
| 366 |
+
|
| 367 |
+
## Best practices
|
| 368 |
+
|
| 369 |
+
1. **Send the system prompt and the control line, exactly as documented.**
|
| 370 |
+
They are the only steering mechanism, and the model was never trained
|
| 371 |
+
without them.
|
| 372 |
+
2. **Decode greedily.** `generation_config.json` already ships `do_sample:
|
| 373 |
+
false`, and for good reason: normalization is a deterministic
|
| 374 |
+
transformation, and sampling only adds variance. If you override the
|
| 375 |
+
config, use temperature 0.
|
| 376 |
+
3. **Size `max_new_tokens` to the input.** Output length closely tracks input
|
| 377 |
+
length; `1.3 × input_tokens + 32` is a safe ceiling, and much cheaper than
|
| 378 |
+
leaving it at 1024.
|
| 379 |
+
4. **Chunk long transcripts at sentence boundaries.** The model is built for
|
| 380 |
+
dictation-length input; keep single passes under roughly 1,000 tokens.
|
| 381 |
+
5. **Expect an empty string sometimes.** Filler-only input returns nothing,
|
| 382 |
+
and your pipeline should treat that as a valid result rather than a
|
| 383 |
+
failure.
|
| 384 |
+
|
| 385 |
+
## Evaluation
|
| 386 |
+
|
| 387 |
+
Evaluated on a held-out English test set of 7,519 cases covering real ASR
|
| 388 |
+
output and synthetic stress sets for numbers, self-corrections, lists, email
|
| 389 |
+
and adversarial inputs: 94.8% token accuracy, measured greedy on the Q4_K_M
|
| 390 |
+
GGUF build. The BF16 weights here should do at least as well.
|
| 391 |
+
|
| 392 |
+
## Using S1-mini in your own app
|
| 393 |
+
|
| 394 |
+
S1-mini is Apache 2.0 plus a naming clause, the same base license it inherits
|
| 395 |
+
from Qwen3-0.6B, so it can be embedded in open-source and commercial software
|
| 396 |
+
alike: dictation apps,
|
| 397 |
+
meeting-notes tools, live captioning, voice-driven editors, or any pipeline
|
| 398 |
+
that has to turn raw ASR output into text a person will read.
|
| 399 |
+
|
| 400 |
+
It is a post-processing stage rather than a standalone system:
|
| 401 |
+
|
| 402 |
+
```
|
| 403 |
+
audio ──▶ ASR (Whisper, Parakeet, …) ──▶ S1-mini ──▶ clean text
|
| 404 |
+
```
|
| 405 |
+
|
| 406 |
+
The ASR's raw transcript becomes the transcript line, your app's settings
|
| 407 |
+
choose the three control-line values, and the model returns text ready to
|
| 408 |
+
display. At 0.6B it is small enough to ship on-device, and the
|
| 409 |
+
[GGUF builds](https://huggingface.co/superwhisper/s1-mini-GGUF) exist for
|
| 410 |
+
exactly that.
|
| 411 |
+
|
| 412 |
+
Nothing about the model is Superwhisper-specific. The two things to get right
|
| 413 |
+
in any integration are the input format documented above and the thinking
|
| 414 |
+
flag; nearly every integration bug traces back to one of those.
|
| 415 |
+
|
| 416 |
+
> [!IMPORTANT]
|
| 417 |
+
> Read the [LICENSE](LICENSE) before you ship. Apache 2.0 is permissive but
|
| 418 |
+
> not obligation-free: you must retain the license text and attribution
|
| 419 |
+
> notices, and state significant changes if you redistribute a modified
|
| 420 |
+
> version. It also carries one additional term: the model must keep its
|
| 421 |
+
> name, "S1-mini" by "Superwhisper", with that exact capitalization,
|
| 422 |
+
> wherever it's used. If you are bundling S1-mini into a commercial
|
| 423 |
+
> dictation app or redistributing the weights yourself, confirm the terms
|
| 424 |
+
> cover your case rather than assuming they do.
|
| 425 |
+
|
| 426 |
+
## License
|
| 427 |
+
|
| 428 |
+
S1-mini is released under Apache 2.0, which it inherits from Qwen3-0.6B, plus
|
| 429 |
+
one additional term: wherever it's used, it must keep its name, "S1-mini" by
|
| 430 |
+
"Superwhisper", with that exact capitalization. See [LICENSE](LICENSE).
|
| 431 |
+
|
| 432 |
+
## Citation
|
| 433 |
+
|
| 434 |
+
```bibtex
|
| 435 |
+
@misc{s1mini2026,
|
| 436 |
+
title = {S1-mini: a small text normalizer for speech-to-text output},
|
| 437 |
+
author = {Superwhisper},
|
| 438 |
+
year = {2026},
|
| 439 |
+
url = {https://huggingface.co/superwhisper/s1-mini}
|
| 440 |
+
}
|
| 441 |
+
```
|
| 442 |
+
|
| 443 |
+
Built on Qwen3:
|
| 444 |
+
|
| 445 |
+
```bibtex
|
| 446 |
+
@misc{qwen3technicalreport,
|
| 447 |
+
title = {Qwen3 Technical Report},
|
| 448 |
+
author = {Qwen Team},
|
| 449 |
+
year = {2025},
|
| 450 |
+
eprint = {2505.09388},
|
| 451 |
+
archivePrefix = {arXiv},
|
| 452 |
+
primaryClass = {cs.CL},
|
| 453 |
+
url = {https://arxiv.org/abs/2505.09388}
|
| 454 |
+
}
|
| 455 |
+
```
|
banner.jpg
ADDED
|
Git LFS Details
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0].role == 'system' %}
|
| 4 |
+
{{- messages[0].content + '\n\n' }}
|
| 5 |
+
{%- endif %}
|
| 6 |
+
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 7 |
+
{%- for tool in tools %}
|
| 8 |
+
{{- "\n" }}
|
| 9 |
+
{{- tool | tojson }}
|
| 10 |
+
{%- endfor %}
|
| 11 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 12 |
+
{%- else %}
|
| 13 |
+
{%- if messages[0].role == 'system' %}
|
| 14 |
+
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
| 15 |
+
{%- endif %}
|
| 16 |
+
{%- endif %}
|
| 17 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 18 |
+
{%- for message in messages[::-1] %}
|
| 19 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 20 |
+
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
| 21 |
+
{%- set ns.multi_step_tool = false %}
|
| 22 |
+
{%- set ns.last_query_index = index %}
|
| 23 |
+
{%- endif %}
|
| 24 |
+
{%- endfor %}
|
| 25 |
+
{%- for message in messages %}
|
| 26 |
+
{%- if message.content is string %}
|
| 27 |
+
{%- set content = message.content %}
|
| 28 |
+
{%- else %}
|
| 29 |
+
{%- set content = '' %}
|
| 30 |
+
{%- endif %}
|
| 31 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
| 32 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 33 |
+
{%- elif message.role == "assistant" %}
|
| 34 |
+
{%- set reasoning_content = '' %}
|
| 35 |
+
{%- if message.reasoning_content is string %}
|
| 36 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 37 |
+
{%- else %}
|
| 38 |
+
{%- if '</think>' in content %}
|
| 39 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 40 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 41 |
+
{%- endif %}
|
| 42 |
+
{%- endif %}
|
| 43 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 44 |
+
{%- if loop.last or (not loop.last and reasoning_content) %}
|
| 45 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
| 46 |
+
{%- else %}
|
| 47 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 48 |
+
{%- endif %}
|
| 49 |
+
{%- else %}
|
| 50 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 51 |
+
{%- endif %}
|
| 52 |
+
{%- if message.tool_calls %}
|
| 53 |
+
{%- for tool_call in message.tool_calls %}
|
| 54 |
+
{%- if (loop.first and content) or (not loop.first) %}
|
| 55 |
+
{{- '\n' }}
|
| 56 |
+
{%- endif %}
|
| 57 |
+
{%- if tool_call.function %}
|
| 58 |
+
{%- set tool_call = tool_call.function %}
|
| 59 |
+
{%- endif %}
|
| 60 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 61 |
+
{{- tool_call.name }}
|
| 62 |
+
{{- '", "arguments": ' }}
|
| 63 |
+
{%- if tool_call.arguments is string %}
|
| 64 |
+
{{- tool_call.arguments }}
|
| 65 |
+
{%- else %}
|
| 66 |
+
{{- tool_call.arguments | tojson }}
|
| 67 |
+
{%- endif %}
|
| 68 |
+
{{- '}\n</tool_call>' }}
|
| 69 |
+
{%- endfor %}
|
| 70 |
+
{%- endif %}
|
| 71 |
+
{{- '<|im_end|>\n' }}
|
| 72 |
+
{%- elif message.role == "tool" %}
|
| 73 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 74 |
+
{{- '<|im_start|>user' }}
|
| 75 |
+
{%- endif %}
|
| 76 |
+
{{- '\n<tool_response>\n' }}
|
| 77 |
+
{{- content }}
|
| 78 |
+
{{- '\n</tool_response>' }}
|
| 79 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 80 |
+
{{- '<|im_end|>\n' }}
|
| 81 |
+
{%- endif %}
|
| 82 |
+
{%- endif %}
|
| 83 |
+
{%- endfor %}
|
| 84 |
+
{%- if add_generation_prompt %}
|
| 85 |
+
{{- '<|im_start|>assistant\n' }}
|
| 86 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 87 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 88 |
+
{%- endif %}
|
| 89 |
+
{%- endif %}
|
config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"eos_token_id": [
|
| 9 |
+
151645,
|
| 10 |
+
151643
|
| 11 |
+
],
|
| 12 |
+
"head_dim": 128,
|
| 13 |
+
"hidden_act": "silu",
|
| 14 |
+
"hidden_size": 1024,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 3072,
|
| 17 |
+
"max_position_embeddings": 40960,
|
| 18 |
+
"max_window_layers": 28,
|
| 19 |
+
"model_type": "qwen3",
|
| 20 |
+
"num_attention_heads": 16,
|
| 21 |
+
"num_hidden_layers": 28,
|
| 22 |
+
"num_key_value_heads": 8,
|
| 23 |
+
"quantization": {
|
| 24 |
+
"group_size": 64,
|
| 25 |
+
"bits": 4,
|
| 26 |
+
"mode": "affine"
|
| 27 |
+
},
|
| 28 |
+
"quantization_config": {
|
| 29 |
+
"group_size": 64,
|
| 30 |
+
"bits": 4,
|
| 31 |
+
"mode": "affine"
|
| 32 |
+
},
|
| 33 |
+
"rms_norm_eps": 1e-06,
|
| 34 |
+
"rope_scaling": null,
|
| 35 |
+
"rope_theta": 1000000,
|
| 36 |
+
"sliding_window": null,
|
| 37 |
+
"tie_word_embeddings": true,
|
| 38 |
+
"torch_dtype": "bfloat16",
|
| 39 |
+
"transformers_version": "4.51.0",
|
| 40 |
+
"use_cache": true,
|
| 41 |
+
"use_sliding_window": false,
|
| 42 |
+
"vocab_size": 151936
|
| 43 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": false,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"transformers_version": "4.51.0"
|
| 10 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:284990f04854426d6164d634d8c472d7eb6e79c68f8fd8b4c8a80a4f234276ba
|
| 3 |
+
size 335450548
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,712 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_size": 335372288,
|
| 4 |
+
"total_parameters": 596049920
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"model.embed_tokens.biases": "model.safetensors",
|
| 8 |
+
"model.embed_tokens.scales": "model.safetensors",
|
| 9 |
+
"model.embed_tokens.weight": "model.safetensors",
|
| 10 |
+
"model.layers.0.input_layernorm.weight": "model.safetensors",
|
| 11 |
+
"model.layers.0.mlp.down_proj.biases": "model.safetensors",
|
| 12 |
+
"model.layers.0.mlp.down_proj.scales": "model.safetensors",
|
| 13 |
+
"model.layers.0.mlp.down_proj.weight": "model.safetensors",
|
| 14 |
+
"model.layers.0.mlp.gate_proj.biases": "model.safetensors",
|
| 15 |
+
"model.layers.0.mlp.gate_proj.scales": "model.safetensors",
|
| 16 |
+
"model.layers.0.mlp.gate_proj.weight": "model.safetensors",
|
| 17 |
+
"model.layers.0.mlp.up_proj.biases": "model.safetensors",
|
| 18 |
+
"model.layers.0.mlp.up_proj.scales": "model.safetensors",
|
| 19 |
+
"model.layers.0.mlp.up_proj.weight": "model.safetensors",
|
| 20 |
+
"model.layers.0.post_attention_layernorm.weight": "model.safetensors",
|
| 21 |
+
"model.layers.0.self_attn.k_norm.weight": "model.safetensors",
|
| 22 |
+
"model.layers.0.self_attn.k_proj.biases": "model.safetensors",
|
| 23 |
+
"model.layers.0.self_attn.k_proj.scales": "model.safetensors",
|
| 24 |
+
"model.layers.0.self_attn.k_proj.weight": "model.safetensors",
|
| 25 |
+
"model.layers.0.self_attn.o_proj.biases": "model.safetensors",
|
| 26 |
+
"model.layers.0.self_attn.o_proj.scales": "model.safetensors",
|
| 27 |
+
"model.layers.0.self_attn.o_proj.weight": "model.safetensors",
|
| 28 |
+
"model.layers.0.self_attn.q_norm.weight": "model.safetensors",
|
| 29 |
+
"model.layers.0.self_attn.q_proj.biases": "model.safetensors",
|
| 30 |
+
"model.layers.0.self_attn.q_proj.scales": "model.safetensors",
|
| 31 |
+
"model.layers.0.self_attn.q_proj.weight": "model.safetensors",
|
| 32 |
+
"model.layers.0.self_attn.v_proj.biases": "model.safetensors",
|
| 33 |
+
"model.layers.0.self_attn.v_proj.scales": "model.safetensors",
|
| 34 |
+
"model.layers.0.self_attn.v_proj.weight": "model.safetensors",
|
| 35 |
+
"model.layers.1.input_layernorm.weight": "model.safetensors",
|
| 36 |
+
"model.layers.1.mlp.down_proj.biases": "model.safetensors",
|
| 37 |
+
"model.layers.1.mlp.down_proj.scales": "model.safetensors",
|
| 38 |
+
"model.layers.1.mlp.down_proj.weight": "model.safetensors",
|
| 39 |
+
"model.layers.1.mlp.gate_proj.biases": "model.safetensors",
|
| 40 |
+
"model.layers.1.mlp.gate_proj.scales": "model.safetensors",
|
| 41 |
+
"model.layers.1.mlp.gate_proj.weight": "model.safetensors",
|
| 42 |
+
"model.layers.1.mlp.up_proj.biases": "model.safetensors",
|
| 43 |
+
"model.layers.1.mlp.up_proj.scales": "model.safetensors",
|
| 44 |
+
"model.layers.1.mlp.up_proj.weight": "model.safetensors",
|
| 45 |
+
"model.layers.1.post_attention_layernorm.weight": "model.safetensors",
|
| 46 |
+
"model.layers.1.self_attn.k_norm.weight": "model.safetensors",
|
| 47 |
+
"model.layers.1.self_attn.k_proj.biases": "model.safetensors",
|
| 48 |
+
"model.layers.1.self_attn.k_proj.scales": "model.safetensors",
|
| 49 |
+
"model.layers.1.self_attn.k_proj.weight": "model.safetensors",
|
| 50 |
+
"model.layers.1.self_attn.o_proj.biases": "model.safetensors",
|
| 51 |
+
"model.layers.1.self_attn.o_proj.scales": "model.safetensors",
|
| 52 |
+
"model.layers.1.self_attn.o_proj.weight": "model.safetensors",
|
| 53 |
+
"model.layers.1.self_attn.q_norm.weight": "model.safetensors",
|
| 54 |
+
"model.layers.1.self_attn.q_proj.biases": "model.safetensors",
|
| 55 |
+
"model.layers.1.self_attn.q_proj.scales": "model.safetensors",
|
| 56 |
+
"model.layers.1.self_attn.q_proj.weight": "model.safetensors",
|
| 57 |
+
"model.layers.1.self_attn.v_proj.biases": "model.safetensors",
|
| 58 |
+
"model.layers.1.self_attn.v_proj.scales": "model.safetensors",
|
| 59 |
+
"model.layers.1.self_attn.v_proj.weight": "model.safetensors",
|
| 60 |
+
"model.layers.10.input_layernorm.weight": "model.safetensors",
|
| 61 |
+
"model.layers.10.mlp.down_proj.biases": "model.safetensors",
|
| 62 |
+
"model.layers.10.mlp.down_proj.scales": "model.safetensors",
|
| 63 |
+
"model.layers.10.mlp.down_proj.weight": "model.safetensors",
|
| 64 |
+
"model.layers.10.mlp.gate_proj.biases": "model.safetensors",
|
| 65 |
+
"model.layers.10.mlp.gate_proj.scales": "model.safetensors",
|
| 66 |
+
"model.layers.10.mlp.gate_proj.weight": "model.safetensors",
|
| 67 |
+
"model.layers.10.mlp.up_proj.biases": "model.safetensors",
|
| 68 |
+
"model.layers.10.mlp.up_proj.scales": "model.safetensors",
|
| 69 |
+
"model.layers.10.mlp.up_proj.weight": "model.safetensors",
|
| 70 |
+
"model.layers.10.post_attention_layernorm.weight": "model.safetensors",
|
| 71 |
+
"model.layers.10.self_attn.k_norm.weight": "model.safetensors",
|
| 72 |
+
"model.layers.10.self_attn.k_proj.biases": "model.safetensors",
|
| 73 |
+
"model.layers.10.self_attn.k_proj.scales": "model.safetensors",
|
| 74 |
+
"model.layers.10.self_attn.k_proj.weight": "model.safetensors",
|
| 75 |
+
"model.layers.10.self_attn.o_proj.biases": "model.safetensors",
|
| 76 |
+
"model.layers.10.self_attn.o_proj.scales": "model.safetensors",
|
| 77 |
+
"model.layers.10.self_attn.o_proj.weight": "model.safetensors",
|
| 78 |
+
"model.layers.10.self_attn.q_norm.weight": "model.safetensors",
|
| 79 |
+
"model.layers.10.self_attn.q_proj.biases": "model.safetensors",
|
| 80 |
+
"model.layers.10.self_attn.q_proj.scales": "model.safetensors",
|
| 81 |
+
"model.layers.10.self_attn.q_proj.weight": "model.safetensors",
|
| 82 |
+
"model.layers.10.self_attn.v_proj.biases": "model.safetensors",
|
| 83 |
+
"model.layers.10.self_attn.v_proj.scales": "model.safetensors",
|
| 84 |
+
"model.layers.10.self_attn.v_proj.weight": "model.safetensors",
|
| 85 |
+
"model.layers.11.input_layernorm.weight": "model.safetensors",
|
| 86 |
+
"model.layers.11.mlp.down_proj.biases": "model.safetensors",
|
| 87 |
+
"model.layers.11.mlp.down_proj.scales": "model.safetensors",
|
| 88 |
+
"model.layers.11.mlp.down_proj.weight": "model.safetensors",
|
| 89 |
+
"model.layers.11.mlp.gate_proj.biases": "model.safetensors",
|
| 90 |
+
"model.layers.11.mlp.gate_proj.scales": "model.safetensors",
|
| 91 |
+
"model.layers.11.mlp.gate_proj.weight": "model.safetensors",
|
| 92 |
+
"model.layers.11.mlp.up_proj.biases": "model.safetensors",
|
| 93 |
+
"model.layers.11.mlp.up_proj.scales": "model.safetensors",
|
| 94 |
+
"model.layers.11.mlp.up_proj.weight": "model.safetensors",
|
| 95 |
+
"model.layers.11.post_attention_layernorm.weight": "model.safetensors",
|
| 96 |
+
"model.layers.11.self_attn.k_norm.weight": "model.safetensors",
|
| 97 |
+
"model.layers.11.self_attn.k_proj.biases": "model.safetensors",
|
| 98 |
+
"model.layers.11.self_attn.k_proj.scales": "model.safetensors",
|
| 99 |
+
"model.layers.11.self_attn.k_proj.weight": "model.safetensors",
|
| 100 |
+
"model.layers.11.self_attn.o_proj.biases": "model.safetensors",
|
| 101 |
+
"model.layers.11.self_attn.o_proj.scales": "model.safetensors",
|
| 102 |
+
"model.layers.11.self_attn.o_proj.weight": "model.safetensors",
|
| 103 |
+
"model.layers.11.self_attn.q_norm.weight": "model.safetensors",
|
| 104 |
+
"model.layers.11.self_attn.q_proj.biases": "model.safetensors",
|
| 105 |
+
"model.layers.11.self_attn.q_proj.scales": "model.safetensors",
|
| 106 |
+
"model.layers.11.self_attn.q_proj.weight": "model.safetensors",
|
| 107 |
+
"model.layers.11.self_attn.v_proj.biases": "model.safetensors",
|
| 108 |
+
"model.layers.11.self_attn.v_proj.scales": "model.safetensors",
|
| 109 |
+
"model.layers.11.self_attn.v_proj.weight": "model.safetensors",
|
| 110 |
+
"model.layers.12.input_layernorm.weight": "model.safetensors",
|
| 111 |
+
"model.layers.12.mlp.down_proj.biases": "model.safetensors",
|
| 112 |
+
"model.layers.12.mlp.down_proj.scales": "model.safetensors",
|
| 113 |
+
"model.layers.12.mlp.down_proj.weight": "model.safetensors",
|
| 114 |
+
"model.layers.12.mlp.gate_proj.biases": "model.safetensors",
|
| 115 |
+
"model.layers.12.mlp.gate_proj.scales": "model.safetensors",
|
| 116 |
+
"model.layers.12.mlp.gate_proj.weight": "model.safetensors",
|
| 117 |
+
"model.layers.12.mlp.up_proj.biases": "model.safetensors",
|
| 118 |
+
"model.layers.12.mlp.up_proj.scales": "model.safetensors",
|
| 119 |
+
"model.layers.12.mlp.up_proj.weight": "model.safetensors",
|
| 120 |
+
"model.layers.12.post_attention_layernorm.weight": "model.safetensors",
|
| 121 |
+
"model.layers.12.self_attn.k_norm.weight": "model.safetensors",
|
| 122 |
+
"model.layers.12.self_attn.k_proj.biases": "model.safetensors",
|
| 123 |
+
"model.layers.12.self_attn.k_proj.scales": "model.safetensors",
|
| 124 |
+
"model.layers.12.self_attn.k_proj.weight": "model.safetensors",
|
| 125 |
+
"model.layers.12.self_attn.o_proj.biases": "model.safetensors",
|
| 126 |
+
"model.layers.12.self_attn.o_proj.scales": "model.safetensors",
|
| 127 |
+
"model.layers.12.self_attn.o_proj.weight": "model.safetensors",
|
| 128 |
+
"model.layers.12.self_attn.q_norm.weight": "model.safetensors",
|
| 129 |
+
"model.layers.12.self_attn.q_proj.biases": "model.safetensors",
|
| 130 |
+
"model.layers.12.self_attn.q_proj.scales": "model.safetensors",
|
| 131 |
+
"model.layers.12.self_attn.q_proj.weight": "model.safetensors",
|
| 132 |
+
"model.layers.12.self_attn.v_proj.biases": "model.safetensors",
|
| 133 |
+
"model.layers.12.self_attn.v_proj.scales": "model.safetensors",
|
| 134 |
+
"model.layers.12.self_attn.v_proj.weight": "model.safetensors",
|
| 135 |
+
"model.layers.13.input_layernorm.weight": "model.safetensors",
|
| 136 |
+
"model.layers.13.mlp.down_proj.biases": "model.safetensors",
|
| 137 |
+
"model.layers.13.mlp.down_proj.scales": "model.safetensors",
|
| 138 |
+
"model.layers.13.mlp.down_proj.weight": "model.safetensors",
|
| 139 |
+
"model.layers.13.mlp.gate_proj.biases": "model.safetensors",
|
| 140 |
+
"model.layers.13.mlp.gate_proj.scales": "model.safetensors",
|
| 141 |
+
"model.layers.13.mlp.gate_proj.weight": "model.safetensors",
|
| 142 |
+
"model.layers.13.mlp.up_proj.biases": "model.safetensors",
|
| 143 |
+
"model.layers.13.mlp.up_proj.scales": "model.safetensors",
|
| 144 |
+
"model.layers.13.mlp.up_proj.weight": "model.safetensors",
|
| 145 |
+
"model.layers.13.post_attention_layernorm.weight": "model.safetensors",
|
| 146 |
+
"model.layers.13.self_attn.k_norm.weight": "model.safetensors",
|
| 147 |
+
"model.layers.13.self_attn.k_proj.biases": "model.safetensors",
|
| 148 |
+
"model.layers.13.self_attn.k_proj.scales": "model.safetensors",
|
| 149 |
+
"model.layers.13.self_attn.k_proj.weight": "model.safetensors",
|
| 150 |
+
"model.layers.13.self_attn.o_proj.biases": "model.safetensors",
|
| 151 |
+
"model.layers.13.self_attn.o_proj.scales": "model.safetensors",
|
| 152 |
+
"model.layers.13.self_attn.o_proj.weight": "model.safetensors",
|
| 153 |
+
"model.layers.13.self_attn.q_norm.weight": "model.safetensors",
|
| 154 |
+
"model.layers.13.self_attn.q_proj.biases": "model.safetensors",
|
| 155 |
+
"model.layers.13.self_attn.q_proj.scales": "model.safetensors",
|
| 156 |
+
"model.layers.13.self_attn.q_proj.weight": "model.safetensors",
|
| 157 |
+
"model.layers.13.self_attn.v_proj.biases": "model.safetensors",
|
| 158 |
+
"model.layers.13.self_attn.v_proj.scales": "model.safetensors",
|
| 159 |
+
"model.layers.13.self_attn.v_proj.weight": "model.safetensors",
|
| 160 |
+
"model.layers.14.input_layernorm.weight": "model.safetensors",
|
| 161 |
+
"model.layers.14.mlp.down_proj.biases": "model.safetensors",
|
| 162 |
+
"model.layers.14.mlp.down_proj.scales": "model.safetensors",
|
| 163 |
+
"model.layers.14.mlp.down_proj.weight": "model.safetensors",
|
| 164 |
+
"model.layers.14.mlp.gate_proj.biases": "model.safetensors",
|
| 165 |
+
"model.layers.14.mlp.gate_proj.scales": "model.safetensors",
|
| 166 |
+
"model.layers.14.mlp.gate_proj.weight": "model.safetensors",
|
| 167 |
+
"model.layers.14.mlp.up_proj.biases": "model.safetensors",
|
| 168 |
+
"model.layers.14.mlp.up_proj.scales": "model.safetensors",
|
| 169 |
+
"model.layers.14.mlp.up_proj.weight": "model.safetensors",
|
| 170 |
+
"model.layers.14.post_attention_layernorm.weight": "model.safetensors",
|
| 171 |
+
"model.layers.14.self_attn.k_norm.weight": "model.safetensors",
|
| 172 |
+
"model.layers.14.self_attn.k_proj.biases": "model.safetensors",
|
| 173 |
+
"model.layers.14.self_attn.k_proj.scales": "model.safetensors",
|
| 174 |
+
"model.layers.14.self_attn.k_proj.weight": "model.safetensors",
|
| 175 |
+
"model.layers.14.self_attn.o_proj.biases": "model.safetensors",
|
| 176 |
+
"model.layers.14.self_attn.o_proj.scales": "model.safetensors",
|
| 177 |
+
"model.layers.14.self_attn.o_proj.weight": "model.safetensors",
|
| 178 |
+
"model.layers.14.self_attn.q_norm.weight": "model.safetensors",
|
| 179 |
+
"model.layers.14.self_attn.q_proj.biases": "model.safetensors",
|
| 180 |
+
"model.layers.14.self_attn.q_proj.scales": "model.safetensors",
|
| 181 |
+
"model.layers.14.self_attn.q_proj.weight": "model.safetensors",
|
| 182 |
+
"model.layers.14.self_attn.v_proj.biases": "model.safetensors",
|
| 183 |
+
"model.layers.14.self_attn.v_proj.scales": "model.safetensors",
|
| 184 |
+
"model.layers.14.self_attn.v_proj.weight": "model.safetensors",
|
| 185 |
+
"model.layers.15.input_layernorm.weight": "model.safetensors",
|
| 186 |
+
"model.layers.15.mlp.down_proj.biases": "model.safetensors",
|
| 187 |
+
"model.layers.15.mlp.down_proj.scales": "model.safetensors",
|
| 188 |
+
"model.layers.15.mlp.down_proj.weight": "model.safetensors",
|
| 189 |
+
"model.layers.15.mlp.gate_proj.biases": "model.safetensors",
|
| 190 |
+
"model.layers.15.mlp.gate_proj.scales": "model.safetensors",
|
| 191 |
+
"model.layers.15.mlp.gate_proj.weight": "model.safetensors",
|
| 192 |
+
"model.layers.15.mlp.up_proj.biases": "model.safetensors",
|
| 193 |
+
"model.layers.15.mlp.up_proj.scales": "model.safetensors",
|
| 194 |
+
"model.layers.15.mlp.up_proj.weight": "model.safetensors",
|
| 195 |
+
"model.layers.15.post_attention_layernorm.weight": "model.safetensors",
|
| 196 |
+
"model.layers.15.self_attn.k_norm.weight": "model.safetensors",
|
| 197 |
+
"model.layers.15.self_attn.k_proj.biases": "model.safetensors",
|
| 198 |
+
"model.layers.15.self_attn.k_proj.scales": "model.safetensors",
|
| 199 |
+
"model.layers.15.self_attn.k_proj.weight": "model.safetensors",
|
| 200 |
+
"model.layers.15.self_attn.o_proj.biases": "model.safetensors",
|
| 201 |
+
"model.layers.15.self_attn.o_proj.scales": "model.safetensors",
|
| 202 |
+
"model.layers.15.self_attn.o_proj.weight": "model.safetensors",
|
| 203 |
+
"model.layers.15.self_attn.q_norm.weight": "model.safetensors",
|
| 204 |
+
"model.layers.15.self_attn.q_proj.biases": "model.safetensors",
|
| 205 |
+
"model.layers.15.self_attn.q_proj.scales": "model.safetensors",
|
| 206 |
+
"model.layers.15.self_attn.q_proj.weight": "model.safetensors",
|
| 207 |
+
"model.layers.15.self_attn.v_proj.biases": "model.safetensors",
|
| 208 |
+
"model.layers.15.self_attn.v_proj.scales": "model.safetensors",
|
| 209 |
+
"model.layers.15.self_attn.v_proj.weight": "model.safetensors",
|
| 210 |
+
"model.layers.16.input_layernorm.weight": "model.safetensors",
|
| 211 |
+
"model.layers.16.mlp.down_proj.biases": "model.safetensors",
|
| 212 |
+
"model.layers.16.mlp.down_proj.scales": "model.safetensors",
|
| 213 |
+
"model.layers.16.mlp.down_proj.weight": "model.safetensors",
|
| 214 |
+
"model.layers.16.mlp.gate_proj.biases": "model.safetensors",
|
| 215 |
+
"model.layers.16.mlp.gate_proj.scales": "model.safetensors",
|
| 216 |
+
"model.layers.16.mlp.gate_proj.weight": "model.safetensors",
|
| 217 |
+
"model.layers.16.mlp.up_proj.biases": "model.safetensors",
|
| 218 |
+
"model.layers.16.mlp.up_proj.scales": "model.safetensors",
|
| 219 |
+
"model.layers.16.mlp.up_proj.weight": "model.safetensors",
|
| 220 |
+
"model.layers.16.post_attention_layernorm.weight": "model.safetensors",
|
| 221 |
+
"model.layers.16.self_attn.k_norm.weight": "model.safetensors",
|
| 222 |
+
"model.layers.16.self_attn.k_proj.biases": "model.safetensors",
|
| 223 |
+
"model.layers.16.self_attn.k_proj.scales": "model.safetensors",
|
| 224 |
+
"model.layers.16.self_attn.k_proj.weight": "model.safetensors",
|
| 225 |
+
"model.layers.16.self_attn.o_proj.biases": "model.safetensors",
|
| 226 |
+
"model.layers.16.self_attn.o_proj.scales": "model.safetensors",
|
| 227 |
+
"model.layers.16.self_attn.o_proj.weight": "model.safetensors",
|
| 228 |
+
"model.layers.16.self_attn.q_norm.weight": "model.safetensors",
|
| 229 |
+
"model.layers.16.self_attn.q_proj.biases": "model.safetensors",
|
| 230 |
+
"model.layers.16.self_attn.q_proj.scales": "model.safetensors",
|
| 231 |
+
"model.layers.16.self_attn.q_proj.weight": "model.safetensors",
|
| 232 |
+
"model.layers.16.self_attn.v_proj.biases": "model.safetensors",
|
| 233 |
+
"model.layers.16.self_attn.v_proj.scales": "model.safetensors",
|
| 234 |
+
"model.layers.16.self_attn.v_proj.weight": "model.safetensors",
|
| 235 |
+
"model.layers.17.input_layernorm.weight": "model.safetensors",
|
| 236 |
+
"model.layers.17.mlp.down_proj.biases": "model.safetensors",
|
| 237 |
+
"model.layers.17.mlp.down_proj.scales": "model.safetensors",
|
| 238 |
+
"model.layers.17.mlp.down_proj.weight": "model.safetensors",
|
| 239 |
+
"model.layers.17.mlp.gate_proj.biases": "model.safetensors",
|
| 240 |
+
"model.layers.17.mlp.gate_proj.scales": "model.safetensors",
|
| 241 |
+
"model.layers.17.mlp.gate_proj.weight": "model.safetensors",
|
| 242 |
+
"model.layers.17.mlp.up_proj.biases": "model.safetensors",
|
| 243 |
+
"model.layers.17.mlp.up_proj.scales": "model.safetensors",
|
| 244 |
+
"model.layers.17.mlp.up_proj.weight": "model.safetensors",
|
| 245 |
+
"model.layers.17.post_attention_layernorm.weight": "model.safetensors",
|
| 246 |
+
"model.layers.17.self_attn.k_norm.weight": "model.safetensors",
|
| 247 |
+
"model.layers.17.self_attn.k_proj.biases": "model.safetensors",
|
| 248 |
+
"model.layers.17.self_attn.k_proj.scales": "model.safetensors",
|
| 249 |
+
"model.layers.17.self_attn.k_proj.weight": "model.safetensors",
|
| 250 |
+
"model.layers.17.self_attn.o_proj.biases": "model.safetensors",
|
| 251 |
+
"model.layers.17.self_attn.o_proj.scales": "model.safetensors",
|
| 252 |
+
"model.layers.17.self_attn.o_proj.weight": "model.safetensors",
|
| 253 |
+
"model.layers.17.self_attn.q_norm.weight": "model.safetensors",
|
| 254 |
+
"model.layers.17.self_attn.q_proj.biases": "model.safetensors",
|
| 255 |
+
"model.layers.17.self_attn.q_proj.scales": "model.safetensors",
|
| 256 |
+
"model.layers.17.self_attn.q_proj.weight": "model.safetensors",
|
| 257 |
+
"model.layers.17.self_attn.v_proj.biases": "model.safetensors",
|
| 258 |
+
"model.layers.17.self_attn.v_proj.scales": "model.safetensors",
|
| 259 |
+
"model.layers.17.self_attn.v_proj.weight": "model.safetensors",
|
| 260 |
+
"model.layers.18.input_layernorm.weight": "model.safetensors",
|
| 261 |
+
"model.layers.18.mlp.down_proj.biases": "model.safetensors",
|
| 262 |
+
"model.layers.18.mlp.down_proj.scales": "model.safetensors",
|
| 263 |
+
"model.layers.18.mlp.down_proj.weight": "model.safetensors",
|
| 264 |
+
"model.layers.18.mlp.gate_proj.biases": "model.safetensors",
|
| 265 |
+
"model.layers.18.mlp.gate_proj.scales": "model.safetensors",
|
| 266 |
+
"model.layers.18.mlp.gate_proj.weight": "model.safetensors",
|
| 267 |
+
"model.layers.18.mlp.up_proj.biases": "model.safetensors",
|
| 268 |
+
"model.layers.18.mlp.up_proj.scales": "model.safetensors",
|
| 269 |
+
"model.layers.18.mlp.up_proj.weight": "model.safetensors",
|
| 270 |
+
"model.layers.18.post_attention_layernorm.weight": "model.safetensors",
|
| 271 |
+
"model.layers.18.self_attn.k_norm.weight": "model.safetensors",
|
| 272 |
+
"model.layers.18.self_attn.k_proj.biases": "model.safetensors",
|
| 273 |
+
"model.layers.18.self_attn.k_proj.scales": "model.safetensors",
|
| 274 |
+
"model.layers.18.self_attn.k_proj.weight": "model.safetensors",
|
| 275 |
+
"model.layers.18.self_attn.o_proj.biases": "model.safetensors",
|
| 276 |
+
"model.layers.18.self_attn.o_proj.scales": "model.safetensors",
|
| 277 |
+
"model.layers.18.self_attn.o_proj.weight": "model.safetensors",
|
| 278 |
+
"model.layers.18.self_attn.q_norm.weight": "model.safetensors",
|
| 279 |
+
"model.layers.18.self_attn.q_proj.biases": "model.safetensors",
|
| 280 |
+
"model.layers.18.self_attn.q_proj.scales": "model.safetensors",
|
| 281 |
+
"model.layers.18.self_attn.q_proj.weight": "model.safetensors",
|
| 282 |
+
"model.layers.18.self_attn.v_proj.biases": "model.safetensors",
|
| 283 |
+
"model.layers.18.self_attn.v_proj.scales": "model.safetensors",
|
| 284 |
+
"model.layers.18.self_attn.v_proj.weight": "model.safetensors",
|
| 285 |
+
"model.layers.19.input_layernorm.weight": "model.safetensors",
|
| 286 |
+
"model.layers.19.mlp.down_proj.biases": "model.safetensors",
|
| 287 |
+
"model.layers.19.mlp.down_proj.scales": "model.safetensors",
|
| 288 |
+
"model.layers.19.mlp.down_proj.weight": "model.safetensors",
|
| 289 |
+
"model.layers.19.mlp.gate_proj.biases": "model.safetensors",
|
| 290 |
+
"model.layers.19.mlp.gate_proj.scales": "model.safetensors",
|
| 291 |
+
"model.layers.19.mlp.gate_proj.weight": "model.safetensors",
|
| 292 |
+
"model.layers.19.mlp.up_proj.biases": "model.safetensors",
|
| 293 |
+
"model.layers.19.mlp.up_proj.scales": "model.safetensors",
|
| 294 |
+
"model.layers.19.mlp.up_proj.weight": "model.safetensors",
|
| 295 |
+
"model.layers.19.post_attention_layernorm.weight": "model.safetensors",
|
| 296 |
+
"model.layers.19.self_attn.k_norm.weight": "model.safetensors",
|
| 297 |
+
"model.layers.19.self_attn.k_proj.biases": "model.safetensors",
|
| 298 |
+
"model.layers.19.self_attn.k_proj.scales": "model.safetensors",
|
| 299 |
+
"model.layers.19.self_attn.k_proj.weight": "model.safetensors",
|
| 300 |
+
"model.layers.19.self_attn.o_proj.biases": "model.safetensors",
|
| 301 |
+
"model.layers.19.self_attn.o_proj.scales": "model.safetensors",
|
| 302 |
+
"model.layers.19.self_attn.o_proj.weight": "model.safetensors",
|
| 303 |
+
"model.layers.19.self_attn.q_norm.weight": "model.safetensors",
|
| 304 |
+
"model.layers.19.self_attn.q_proj.biases": "model.safetensors",
|
| 305 |
+
"model.layers.19.self_attn.q_proj.scales": "model.safetensors",
|
| 306 |
+
"model.layers.19.self_attn.q_proj.weight": "model.safetensors",
|
| 307 |
+
"model.layers.19.self_attn.v_proj.biases": "model.safetensors",
|
| 308 |
+
"model.layers.19.self_attn.v_proj.scales": "model.safetensors",
|
| 309 |
+
"model.layers.19.self_attn.v_proj.weight": "model.safetensors",
|
| 310 |
+
"model.layers.2.input_layernorm.weight": "model.safetensors",
|
| 311 |
+
"model.layers.2.mlp.down_proj.biases": "model.safetensors",
|
| 312 |
+
"model.layers.2.mlp.down_proj.scales": "model.safetensors",
|
| 313 |
+
"model.layers.2.mlp.down_proj.weight": "model.safetensors",
|
| 314 |
+
"model.layers.2.mlp.gate_proj.biases": "model.safetensors",
|
| 315 |
+
"model.layers.2.mlp.gate_proj.scales": "model.safetensors",
|
| 316 |
+
"model.layers.2.mlp.gate_proj.weight": "model.safetensors",
|
| 317 |
+
"model.layers.2.mlp.up_proj.biases": "model.safetensors",
|
| 318 |
+
"model.layers.2.mlp.up_proj.scales": "model.safetensors",
|
| 319 |
+
"model.layers.2.mlp.up_proj.weight": "model.safetensors",
|
| 320 |
+
"model.layers.2.post_attention_layernorm.weight": "model.safetensors",
|
| 321 |
+
"model.layers.2.self_attn.k_norm.weight": "model.safetensors",
|
| 322 |
+
"model.layers.2.self_attn.k_proj.biases": "model.safetensors",
|
| 323 |
+
"model.layers.2.self_attn.k_proj.scales": "model.safetensors",
|
| 324 |
+
"model.layers.2.self_attn.k_proj.weight": "model.safetensors",
|
| 325 |
+
"model.layers.2.self_attn.o_proj.biases": "model.safetensors",
|
| 326 |
+
"model.layers.2.self_attn.o_proj.scales": "model.safetensors",
|
| 327 |
+
"model.layers.2.self_attn.o_proj.weight": "model.safetensors",
|
| 328 |
+
"model.layers.2.self_attn.q_norm.weight": "model.safetensors",
|
| 329 |
+
"model.layers.2.self_attn.q_proj.biases": "model.safetensors",
|
| 330 |
+
"model.layers.2.self_attn.q_proj.scales": "model.safetensors",
|
| 331 |
+
"model.layers.2.self_attn.q_proj.weight": "model.safetensors",
|
| 332 |
+
"model.layers.2.self_attn.v_proj.biases": "model.safetensors",
|
| 333 |
+
"model.layers.2.self_attn.v_proj.scales": "model.safetensors",
|
| 334 |
+
"model.layers.2.self_attn.v_proj.weight": "model.safetensors",
|
| 335 |
+
"model.layers.20.input_layernorm.weight": "model.safetensors",
|
| 336 |
+
"model.layers.20.mlp.down_proj.biases": "model.safetensors",
|
| 337 |
+
"model.layers.20.mlp.down_proj.scales": "model.safetensors",
|
| 338 |
+
"model.layers.20.mlp.down_proj.weight": "model.safetensors",
|
| 339 |
+
"model.layers.20.mlp.gate_proj.biases": "model.safetensors",
|
| 340 |
+
"model.layers.20.mlp.gate_proj.scales": "model.safetensors",
|
| 341 |
+
"model.layers.20.mlp.gate_proj.weight": "model.safetensors",
|
| 342 |
+
"model.layers.20.mlp.up_proj.biases": "model.safetensors",
|
| 343 |
+
"model.layers.20.mlp.up_proj.scales": "model.safetensors",
|
| 344 |
+
"model.layers.20.mlp.up_proj.weight": "model.safetensors",
|
| 345 |
+
"model.layers.20.post_attention_layernorm.weight": "model.safetensors",
|
| 346 |
+
"model.layers.20.self_attn.k_norm.weight": "model.safetensors",
|
| 347 |
+
"model.layers.20.self_attn.k_proj.biases": "model.safetensors",
|
| 348 |
+
"model.layers.20.self_attn.k_proj.scales": "model.safetensors",
|
| 349 |
+
"model.layers.20.self_attn.k_proj.weight": "model.safetensors",
|
| 350 |
+
"model.layers.20.self_attn.o_proj.biases": "model.safetensors",
|
| 351 |
+
"model.layers.20.self_attn.o_proj.scales": "model.safetensors",
|
| 352 |
+
"model.layers.20.self_attn.o_proj.weight": "model.safetensors",
|
| 353 |
+
"model.layers.20.self_attn.q_norm.weight": "model.safetensors",
|
| 354 |
+
"model.layers.20.self_attn.q_proj.biases": "model.safetensors",
|
| 355 |
+
"model.layers.20.self_attn.q_proj.scales": "model.safetensors",
|
| 356 |
+
"model.layers.20.self_attn.q_proj.weight": "model.safetensors",
|
| 357 |
+
"model.layers.20.self_attn.v_proj.biases": "model.safetensors",
|
| 358 |
+
"model.layers.20.self_attn.v_proj.scales": "model.safetensors",
|
| 359 |
+
"model.layers.20.self_attn.v_proj.weight": "model.safetensors",
|
| 360 |
+
"model.layers.21.input_layernorm.weight": "model.safetensors",
|
| 361 |
+
"model.layers.21.mlp.down_proj.biases": "model.safetensors",
|
| 362 |
+
"model.layers.21.mlp.down_proj.scales": "model.safetensors",
|
| 363 |
+
"model.layers.21.mlp.down_proj.weight": "model.safetensors",
|
| 364 |
+
"model.layers.21.mlp.gate_proj.biases": "model.safetensors",
|
| 365 |
+
"model.layers.21.mlp.gate_proj.scales": "model.safetensors",
|
| 366 |
+
"model.layers.21.mlp.gate_proj.weight": "model.safetensors",
|
| 367 |
+
"model.layers.21.mlp.up_proj.biases": "model.safetensors",
|
| 368 |
+
"model.layers.21.mlp.up_proj.scales": "model.safetensors",
|
| 369 |
+
"model.layers.21.mlp.up_proj.weight": "model.safetensors",
|
| 370 |
+
"model.layers.21.post_attention_layernorm.weight": "model.safetensors",
|
| 371 |
+
"model.layers.21.self_attn.k_norm.weight": "model.safetensors",
|
| 372 |
+
"model.layers.21.self_attn.k_proj.biases": "model.safetensors",
|
| 373 |
+
"model.layers.21.self_attn.k_proj.scales": "model.safetensors",
|
| 374 |
+
"model.layers.21.self_attn.k_proj.weight": "model.safetensors",
|
| 375 |
+
"model.layers.21.self_attn.o_proj.biases": "model.safetensors",
|
| 376 |
+
"model.layers.21.self_attn.o_proj.scales": "model.safetensors",
|
| 377 |
+
"model.layers.21.self_attn.o_proj.weight": "model.safetensors",
|
| 378 |
+
"model.layers.21.self_attn.q_norm.weight": "model.safetensors",
|
| 379 |
+
"model.layers.21.self_attn.q_proj.biases": "model.safetensors",
|
| 380 |
+
"model.layers.21.self_attn.q_proj.scales": "model.safetensors",
|
| 381 |
+
"model.layers.21.self_attn.q_proj.weight": "model.safetensors",
|
| 382 |
+
"model.layers.21.self_attn.v_proj.biases": "model.safetensors",
|
| 383 |
+
"model.layers.21.self_attn.v_proj.scales": "model.safetensors",
|
| 384 |
+
"model.layers.21.self_attn.v_proj.weight": "model.safetensors",
|
| 385 |
+
"model.layers.22.input_layernorm.weight": "model.safetensors",
|
| 386 |
+
"model.layers.22.mlp.down_proj.biases": "model.safetensors",
|
| 387 |
+
"model.layers.22.mlp.down_proj.scales": "model.safetensors",
|
| 388 |
+
"model.layers.22.mlp.down_proj.weight": "model.safetensors",
|
| 389 |
+
"model.layers.22.mlp.gate_proj.biases": "model.safetensors",
|
| 390 |
+
"model.layers.22.mlp.gate_proj.scales": "model.safetensors",
|
| 391 |
+
"model.layers.22.mlp.gate_proj.weight": "model.safetensors",
|
| 392 |
+
"model.layers.22.mlp.up_proj.biases": "model.safetensors",
|
| 393 |
+
"model.layers.22.mlp.up_proj.scales": "model.safetensors",
|
| 394 |
+
"model.layers.22.mlp.up_proj.weight": "model.safetensors",
|
| 395 |
+
"model.layers.22.post_attention_layernorm.weight": "model.safetensors",
|
| 396 |
+
"model.layers.22.self_attn.k_norm.weight": "model.safetensors",
|
| 397 |
+
"model.layers.22.self_attn.k_proj.biases": "model.safetensors",
|
| 398 |
+
"model.layers.22.self_attn.k_proj.scales": "model.safetensors",
|
| 399 |
+
"model.layers.22.self_attn.k_proj.weight": "model.safetensors",
|
| 400 |
+
"model.layers.22.self_attn.o_proj.biases": "model.safetensors",
|
| 401 |
+
"model.layers.22.self_attn.o_proj.scales": "model.safetensors",
|
| 402 |
+
"model.layers.22.self_attn.o_proj.weight": "model.safetensors",
|
| 403 |
+
"model.layers.22.self_attn.q_norm.weight": "model.safetensors",
|
| 404 |
+
"model.layers.22.self_attn.q_proj.biases": "model.safetensors",
|
| 405 |
+
"model.layers.22.self_attn.q_proj.scales": "model.safetensors",
|
| 406 |
+
"model.layers.22.self_attn.q_proj.weight": "model.safetensors",
|
| 407 |
+
"model.layers.22.self_attn.v_proj.biases": "model.safetensors",
|
| 408 |
+
"model.layers.22.self_attn.v_proj.scales": "model.safetensors",
|
| 409 |
+
"model.layers.22.self_attn.v_proj.weight": "model.safetensors",
|
| 410 |
+
"model.layers.23.input_layernorm.weight": "model.safetensors",
|
| 411 |
+
"model.layers.23.mlp.down_proj.biases": "model.safetensors",
|
| 412 |
+
"model.layers.23.mlp.down_proj.scales": "model.safetensors",
|
| 413 |
+
"model.layers.23.mlp.down_proj.weight": "model.safetensors",
|
| 414 |
+
"model.layers.23.mlp.gate_proj.biases": "model.safetensors",
|
| 415 |
+
"model.layers.23.mlp.gate_proj.scales": "model.safetensors",
|
| 416 |
+
"model.layers.23.mlp.gate_proj.weight": "model.safetensors",
|
| 417 |
+
"model.layers.23.mlp.up_proj.biases": "model.safetensors",
|
| 418 |
+
"model.layers.23.mlp.up_proj.scales": "model.safetensors",
|
| 419 |
+
"model.layers.23.mlp.up_proj.weight": "model.safetensors",
|
| 420 |
+
"model.layers.23.post_attention_layernorm.weight": "model.safetensors",
|
| 421 |
+
"model.layers.23.self_attn.k_norm.weight": "model.safetensors",
|
| 422 |
+
"model.layers.23.self_attn.k_proj.biases": "model.safetensors",
|
| 423 |
+
"model.layers.23.self_attn.k_proj.scales": "model.safetensors",
|
| 424 |
+
"model.layers.23.self_attn.k_proj.weight": "model.safetensors",
|
| 425 |
+
"model.layers.23.self_attn.o_proj.biases": "model.safetensors",
|
| 426 |
+
"model.layers.23.self_attn.o_proj.scales": "model.safetensors",
|
| 427 |
+
"model.layers.23.self_attn.o_proj.weight": "model.safetensors",
|
| 428 |
+
"model.layers.23.self_attn.q_norm.weight": "model.safetensors",
|
| 429 |
+
"model.layers.23.self_attn.q_proj.biases": "model.safetensors",
|
| 430 |
+
"model.layers.23.self_attn.q_proj.scales": "model.safetensors",
|
| 431 |
+
"model.layers.23.self_attn.q_proj.weight": "model.safetensors",
|
| 432 |
+
"model.layers.23.self_attn.v_proj.biases": "model.safetensors",
|
| 433 |
+
"model.layers.23.self_attn.v_proj.scales": "model.safetensors",
|
| 434 |
+
"model.layers.23.self_attn.v_proj.weight": "model.safetensors",
|
| 435 |
+
"model.layers.24.input_layernorm.weight": "model.safetensors",
|
| 436 |
+
"model.layers.24.mlp.down_proj.biases": "model.safetensors",
|
| 437 |
+
"model.layers.24.mlp.down_proj.scales": "model.safetensors",
|
| 438 |
+
"model.layers.24.mlp.down_proj.weight": "model.safetensors",
|
| 439 |
+
"model.layers.24.mlp.gate_proj.biases": "model.safetensors",
|
| 440 |
+
"model.layers.24.mlp.gate_proj.scales": "model.safetensors",
|
| 441 |
+
"model.layers.24.mlp.gate_proj.weight": "model.safetensors",
|
| 442 |
+
"model.layers.24.mlp.up_proj.biases": "model.safetensors",
|
| 443 |
+
"model.layers.24.mlp.up_proj.scales": "model.safetensors",
|
| 444 |
+
"model.layers.24.mlp.up_proj.weight": "model.safetensors",
|
| 445 |
+
"model.layers.24.post_attention_layernorm.weight": "model.safetensors",
|
| 446 |
+
"model.layers.24.self_attn.k_norm.weight": "model.safetensors",
|
| 447 |
+
"model.layers.24.self_attn.k_proj.biases": "model.safetensors",
|
| 448 |
+
"model.layers.24.self_attn.k_proj.scales": "model.safetensors",
|
| 449 |
+
"model.layers.24.self_attn.k_proj.weight": "model.safetensors",
|
| 450 |
+
"model.layers.24.self_attn.o_proj.biases": "model.safetensors",
|
| 451 |
+
"model.layers.24.self_attn.o_proj.scales": "model.safetensors",
|
| 452 |
+
"model.layers.24.self_attn.o_proj.weight": "model.safetensors",
|
| 453 |
+
"model.layers.24.self_attn.q_norm.weight": "model.safetensors",
|
| 454 |
+
"model.layers.24.self_attn.q_proj.biases": "model.safetensors",
|
| 455 |
+
"model.layers.24.self_attn.q_proj.scales": "model.safetensors",
|
| 456 |
+
"model.layers.24.self_attn.q_proj.weight": "model.safetensors",
|
| 457 |
+
"model.layers.24.self_attn.v_proj.biases": "model.safetensors",
|
| 458 |
+
"model.layers.24.self_attn.v_proj.scales": "model.safetensors",
|
| 459 |
+
"model.layers.24.self_attn.v_proj.weight": "model.safetensors",
|
| 460 |
+
"model.layers.25.input_layernorm.weight": "model.safetensors",
|
| 461 |
+
"model.layers.25.mlp.down_proj.biases": "model.safetensors",
|
| 462 |
+
"model.layers.25.mlp.down_proj.scales": "model.safetensors",
|
| 463 |
+
"model.layers.25.mlp.down_proj.weight": "model.safetensors",
|
| 464 |
+
"model.layers.25.mlp.gate_proj.biases": "model.safetensors",
|
| 465 |
+
"model.layers.25.mlp.gate_proj.scales": "model.safetensors",
|
| 466 |
+
"model.layers.25.mlp.gate_proj.weight": "model.safetensors",
|
| 467 |
+
"model.layers.25.mlp.up_proj.biases": "model.safetensors",
|
| 468 |
+
"model.layers.25.mlp.up_proj.scales": "model.safetensors",
|
| 469 |
+
"model.layers.25.mlp.up_proj.weight": "model.safetensors",
|
| 470 |
+
"model.layers.25.post_attention_layernorm.weight": "model.safetensors",
|
| 471 |
+
"model.layers.25.self_attn.k_norm.weight": "model.safetensors",
|
| 472 |
+
"model.layers.25.self_attn.k_proj.biases": "model.safetensors",
|
| 473 |
+
"model.layers.25.self_attn.k_proj.scales": "model.safetensors",
|
| 474 |
+
"model.layers.25.self_attn.k_proj.weight": "model.safetensors",
|
| 475 |
+
"model.layers.25.self_attn.o_proj.biases": "model.safetensors",
|
| 476 |
+
"model.layers.25.self_attn.o_proj.scales": "model.safetensors",
|
| 477 |
+
"model.layers.25.self_attn.o_proj.weight": "model.safetensors",
|
| 478 |
+
"model.layers.25.self_attn.q_norm.weight": "model.safetensors",
|
| 479 |
+
"model.layers.25.self_attn.q_proj.biases": "model.safetensors",
|
| 480 |
+
"model.layers.25.self_attn.q_proj.scales": "model.safetensors",
|
| 481 |
+
"model.layers.25.self_attn.q_proj.weight": "model.safetensors",
|
| 482 |
+
"model.layers.25.self_attn.v_proj.biases": "model.safetensors",
|
| 483 |
+
"model.layers.25.self_attn.v_proj.scales": "model.safetensors",
|
| 484 |
+
"model.layers.25.self_attn.v_proj.weight": "model.safetensors",
|
| 485 |
+
"model.layers.26.input_layernorm.weight": "model.safetensors",
|
| 486 |
+
"model.layers.26.mlp.down_proj.biases": "model.safetensors",
|
| 487 |
+
"model.layers.26.mlp.down_proj.scales": "model.safetensors",
|
| 488 |
+
"model.layers.26.mlp.down_proj.weight": "model.safetensors",
|
| 489 |
+
"model.layers.26.mlp.gate_proj.biases": "model.safetensors",
|
| 490 |
+
"model.layers.26.mlp.gate_proj.scales": "model.safetensors",
|
| 491 |
+
"model.layers.26.mlp.gate_proj.weight": "model.safetensors",
|
| 492 |
+
"model.layers.26.mlp.up_proj.biases": "model.safetensors",
|
| 493 |
+
"model.layers.26.mlp.up_proj.scales": "model.safetensors",
|
| 494 |
+
"model.layers.26.mlp.up_proj.weight": "model.safetensors",
|
| 495 |
+
"model.layers.26.post_attention_layernorm.weight": "model.safetensors",
|
| 496 |
+
"model.layers.26.self_attn.k_norm.weight": "model.safetensors",
|
| 497 |
+
"model.layers.26.self_attn.k_proj.biases": "model.safetensors",
|
| 498 |
+
"model.layers.26.self_attn.k_proj.scales": "model.safetensors",
|
| 499 |
+
"model.layers.26.self_attn.k_proj.weight": "model.safetensors",
|
| 500 |
+
"model.layers.26.self_attn.o_proj.biases": "model.safetensors",
|
| 501 |
+
"model.layers.26.self_attn.o_proj.scales": "model.safetensors",
|
| 502 |
+
"model.layers.26.self_attn.o_proj.weight": "model.safetensors",
|
| 503 |
+
"model.layers.26.self_attn.q_norm.weight": "model.safetensors",
|
| 504 |
+
"model.layers.26.self_attn.q_proj.biases": "model.safetensors",
|
| 505 |
+
"model.layers.26.self_attn.q_proj.scales": "model.safetensors",
|
| 506 |
+
"model.layers.26.self_attn.q_proj.weight": "model.safetensors",
|
| 507 |
+
"model.layers.26.self_attn.v_proj.biases": "model.safetensors",
|
| 508 |
+
"model.layers.26.self_attn.v_proj.scales": "model.safetensors",
|
| 509 |
+
"model.layers.26.self_attn.v_proj.weight": "model.safetensors",
|
| 510 |
+
"model.layers.27.input_layernorm.weight": "model.safetensors",
|
| 511 |
+
"model.layers.27.mlp.down_proj.biases": "model.safetensors",
|
| 512 |
+
"model.layers.27.mlp.down_proj.scales": "model.safetensors",
|
| 513 |
+
"model.layers.27.mlp.down_proj.weight": "model.safetensors",
|
| 514 |
+
"model.layers.27.mlp.gate_proj.biases": "model.safetensors",
|
| 515 |
+
"model.layers.27.mlp.gate_proj.scales": "model.safetensors",
|
| 516 |
+
"model.layers.27.mlp.gate_proj.weight": "model.safetensors",
|
| 517 |
+
"model.layers.27.mlp.up_proj.biases": "model.safetensors",
|
| 518 |
+
"model.layers.27.mlp.up_proj.scales": "model.safetensors",
|
| 519 |
+
"model.layers.27.mlp.up_proj.weight": "model.safetensors",
|
| 520 |
+
"model.layers.27.post_attention_layernorm.weight": "model.safetensors",
|
| 521 |
+
"model.layers.27.self_attn.k_norm.weight": "model.safetensors",
|
| 522 |
+
"model.layers.27.self_attn.k_proj.biases": "model.safetensors",
|
| 523 |
+
"model.layers.27.self_attn.k_proj.scales": "model.safetensors",
|
| 524 |
+
"model.layers.27.self_attn.k_proj.weight": "model.safetensors",
|
| 525 |
+
"model.layers.27.self_attn.o_proj.biases": "model.safetensors",
|
| 526 |
+
"model.layers.27.self_attn.o_proj.scales": "model.safetensors",
|
| 527 |
+
"model.layers.27.self_attn.o_proj.weight": "model.safetensors",
|
| 528 |
+
"model.layers.27.self_attn.q_norm.weight": "model.safetensors",
|
| 529 |
+
"model.layers.27.self_attn.q_proj.biases": "model.safetensors",
|
| 530 |
+
"model.layers.27.self_attn.q_proj.scales": "model.safetensors",
|
| 531 |
+
"model.layers.27.self_attn.q_proj.weight": "model.safetensors",
|
| 532 |
+
"model.layers.27.self_attn.v_proj.biases": "model.safetensors",
|
| 533 |
+
"model.layers.27.self_attn.v_proj.scales": "model.safetensors",
|
| 534 |
+
"model.layers.27.self_attn.v_proj.weight": "model.safetensors",
|
| 535 |
+
"model.layers.3.input_layernorm.weight": "model.safetensors",
|
| 536 |
+
"model.layers.3.mlp.down_proj.biases": "model.safetensors",
|
| 537 |
+
"model.layers.3.mlp.down_proj.scales": "model.safetensors",
|
| 538 |
+
"model.layers.3.mlp.down_proj.weight": "model.safetensors",
|
| 539 |
+
"model.layers.3.mlp.gate_proj.biases": "model.safetensors",
|
| 540 |
+
"model.layers.3.mlp.gate_proj.scales": "model.safetensors",
|
| 541 |
+
"model.layers.3.mlp.gate_proj.weight": "model.safetensors",
|
| 542 |
+
"model.layers.3.mlp.up_proj.biases": "model.safetensors",
|
| 543 |
+
"model.layers.3.mlp.up_proj.scales": "model.safetensors",
|
| 544 |
+
"model.layers.3.mlp.up_proj.weight": "model.safetensors",
|
| 545 |
+
"model.layers.3.post_attention_layernorm.weight": "model.safetensors",
|
| 546 |
+
"model.layers.3.self_attn.k_norm.weight": "model.safetensors",
|
| 547 |
+
"model.layers.3.self_attn.k_proj.biases": "model.safetensors",
|
| 548 |
+
"model.layers.3.self_attn.k_proj.scales": "model.safetensors",
|
| 549 |
+
"model.layers.3.self_attn.k_proj.weight": "model.safetensors",
|
| 550 |
+
"model.layers.3.self_attn.o_proj.biases": "model.safetensors",
|
| 551 |
+
"model.layers.3.self_attn.o_proj.scales": "model.safetensors",
|
| 552 |
+
"model.layers.3.self_attn.o_proj.weight": "model.safetensors",
|
| 553 |
+
"model.layers.3.self_attn.q_norm.weight": "model.safetensors",
|
| 554 |
+
"model.layers.3.self_attn.q_proj.biases": "model.safetensors",
|
| 555 |
+
"model.layers.3.self_attn.q_proj.scales": "model.safetensors",
|
| 556 |
+
"model.layers.3.self_attn.q_proj.weight": "model.safetensors",
|
| 557 |
+
"model.layers.3.self_attn.v_proj.biases": "model.safetensors",
|
| 558 |
+
"model.layers.3.self_attn.v_proj.scales": "model.safetensors",
|
| 559 |
+
"model.layers.3.self_attn.v_proj.weight": "model.safetensors",
|
| 560 |
+
"model.layers.4.input_layernorm.weight": "model.safetensors",
|
| 561 |
+
"model.layers.4.mlp.down_proj.biases": "model.safetensors",
|
| 562 |
+
"model.layers.4.mlp.down_proj.scales": "model.safetensors",
|
| 563 |
+
"model.layers.4.mlp.down_proj.weight": "model.safetensors",
|
| 564 |
+
"model.layers.4.mlp.gate_proj.biases": "model.safetensors",
|
| 565 |
+
"model.layers.4.mlp.gate_proj.scales": "model.safetensors",
|
| 566 |
+
"model.layers.4.mlp.gate_proj.weight": "model.safetensors",
|
| 567 |
+
"model.layers.4.mlp.up_proj.biases": "model.safetensors",
|
| 568 |
+
"model.layers.4.mlp.up_proj.scales": "model.safetensors",
|
| 569 |
+
"model.layers.4.mlp.up_proj.weight": "model.safetensors",
|
| 570 |
+
"model.layers.4.post_attention_layernorm.weight": "model.safetensors",
|
| 571 |
+
"model.layers.4.self_attn.k_norm.weight": "model.safetensors",
|
| 572 |
+
"model.layers.4.self_attn.k_proj.biases": "model.safetensors",
|
| 573 |
+
"model.layers.4.self_attn.k_proj.scales": "model.safetensors",
|
| 574 |
+
"model.layers.4.self_attn.k_proj.weight": "model.safetensors",
|
| 575 |
+
"model.layers.4.self_attn.o_proj.biases": "model.safetensors",
|
| 576 |
+
"model.layers.4.self_attn.o_proj.scales": "model.safetensors",
|
| 577 |
+
"model.layers.4.self_attn.o_proj.weight": "model.safetensors",
|
| 578 |
+
"model.layers.4.self_attn.q_norm.weight": "model.safetensors",
|
| 579 |
+
"model.layers.4.self_attn.q_proj.biases": "model.safetensors",
|
| 580 |
+
"model.layers.4.self_attn.q_proj.scales": "model.safetensors",
|
| 581 |
+
"model.layers.4.self_attn.q_proj.weight": "model.safetensors",
|
| 582 |
+
"model.layers.4.self_attn.v_proj.biases": "model.safetensors",
|
| 583 |
+
"model.layers.4.self_attn.v_proj.scales": "model.safetensors",
|
| 584 |
+
"model.layers.4.self_attn.v_proj.weight": "model.safetensors",
|
| 585 |
+
"model.layers.5.input_layernorm.weight": "model.safetensors",
|
| 586 |
+
"model.layers.5.mlp.down_proj.biases": "model.safetensors",
|
| 587 |
+
"model.layers.5.mlp.down_proj.scales": "model.safetensors",
|
| 588 |
+
"model.layers.5.mlp.down_proj.weight": "model.safetensors",
|
| 589 |
+
"model.layers.5.mlp.gate_proj.biases": "model.safetensors",
|
| 590 |
+
"model.layers.5.mlp.gate_proj.scales": "model.safetensors",
|
| 591 |
+
"model.layers.5.mlp.gate_proj.weight": "model.safetensors",
|
| 592 |
+
"model.layers.5.mlp.up_proj.biases": "model.safetensors",
|
| 593 |
+
"model.layers.5.mlp.up_proj.scales": "model.safetensors",
|
| 594 |
+
"model.layers.5.mlp.up_proj.weight": "model.safetensors",
|
| 595 |
+
"model.layers.5.post_attention_layernorm.weight": "model.safetensors",
|
| 596 |
+
"model.layers.5.self_attn.k_norm.weight": "model.safetensors",
|
| 597 |
+
"model.layers.5.self_attn.k_proj.biases": "model.safetensors",
|
| 598 |
+
"model.layers.5.self_attn.k_proj.scales": "model.safetensors",
|
| 599 |
+
"model.layers.5.self_attn.k_proj.weight": "model.safetensors",
|
| 600 |
+
"model.layers.5.self_attn.o_proj.biases": "model.safetensors",
|
| 601 |
+
"model.layers.5.self_attn.o_proj.scales": "model.safetensors",
|
| 602 |
+
"model.layers.5.self_attn.o_proj.weight": "model.safetensors",
|
| 603 |
+
"model.layers.5.self_attn.q_norm.weight": "model.safetensors",
|
| 604 |
+
"model.layers.5.self_attn.q_proj.biases": "model.safetensors",
|
| 605 |
+
"model.layers.5.self_attn.q_proj.scales": "model.safetensors",
|
| 606 |
+
"model.layers.5.self_attn.q_proj.weight": "model.safetensors",
|
| 607 |
+
"model.layers.5.self_attn.v_proj.biases": "model.safetensors",
|
| 608 |
+
"model.layers.5.self_attn.v_proj.scales": "model.safetensors",
|
| 609 |
+
"model.layers.5.self_attn.v_proj.weight": "model.safetensors",
|
| 610 |
+
"model.layers.6.input_layernorm.weight": "model.safetensors",
|
| 611 |
+
"model.layers.6.mlp.down_proj.biases": "model.safetensors",
|
| 612 |
+
"model.layers.6.mlp.down_proj.scales": "model.safetensors",
|
| 613 |
+
"model.layers.6.mlp.down_proj.weight": "model.safetensors",
|
| 614 |
+
"model.layers.6.mlp.gate_proj.biases": "model.safetensors",
|
| 615 |
+
"model.layers.6.mlp.gate_proj.scales": "model.safetensors",
|
| 616 |
+
"model.layers.6.mlp.gate_proj.weight": "model.safetensors",
|
| 617 |
+
"model.layers.6.mlp.up_proj.biases": "model.safetensors",
|
| 618 |
+
"model.layers.6.mlp.up_proj.scales": "model.safetensors",
|
| 619 |
+
"model.layers.6.mlp.up_proj.weight": "model.safetensors",
|
| 620 |
+
"model.layers.6.post_attention_layernorm.weight": "model.safetensors",
|
| 621 |
+
"model.layers.6.self_attn.k_norm.weight": "model.safetensors",
|
| 622 |
+
"model.layers.6.self_attn.k_proj.biases": "model.safetensors",
|
| 623 |
+
"model.layers.6.self_attn.k_proj.scales": "model.safetensors",
|
| 624 |
+
"model.layers.6.self_attn.k_proj.weight": "model.safetensors",
|
| 625 |
+
"model.layers.6.self_attn.o_proj.biases": "model.safetensors",
|
| 626 |
+
"model.layers.6.self_attn.o_proj.scales": "model.safetensors",
|
| 627 |
+
"model.layers.6.self_attn.o_proj.weight": "model.safetensors",
|
| 628 |
+
"model.layers.6.self_attn.q_norm.weight": "model.safetensors",
|
| 629 |
+
"model.layers.6.self_attn.q_proj.biases": "model.safetensors",
|
| 630 |
+
"model.layers.6.self_attn.q_proj.scales": "model.safetensors",
|
| 631 |
+
"model.layers.6.self_attn.q_proj.weight": "model.safetensors",
|
| 632 |
+
"model.layers.6.self_attn.v_proj.biases": "model.safetensors",
|
| 633 |
+
"model.layers.6.self_attn.v_proj.scales": "model.safetensors",
|
| 634 |
+
"model.layers.6.self_attn.v_proj.weight": "model.safetensors",
|
| 635 |
+
"model.layers.7.input_layernorm.weight": "model.safetensors",
|
| 636 |
+
"model.layers.7.mlp.down_proj.biases": "model.safetensors",
|
| 637 |
+
"model.layers.7.mlp.down_proj.scales": "model.safetensors",
|
| 638 |
+
"model.layers.7.mlp.down_proj.weight": "model.safetensors",
|
| 639 |
+
"model.layers.7.mlp.gate_proj.biases": "model.safetensors",
|
| 640 |
+
"model.layers.7.mlp.gate_proj.scales": "model.safetensors",
|
| 641 |
+
"model.layers.7.mlp.gate_proj.weight": "model.safetensors",
|
| 642 |
+
"model.layers.7.mlp.up_proj.biases": "model.safetensors",
|
| 643 |
+
"model.layers.7.mlp.up_proj.scales": "model.safetensors",
|
| 644 |
+
"model.layers.7.mlp.up_proj.weight": "model.safetensors",
|
| 645 |
+
"model.layers.7.post_attention_layernorm.weight": "model.safetensors",
|
| 646 |
+
"model.layers.7.self_attn.k_norm.weight": "model.safetensors",
|
| 647 |
+
"model.layers.7.self_attn.k_proj.biases": "model.safetensors",
|
| 648 |
+
"model.layers.7.self_attn.k_proj.scales": "model.safetensors",
|
| 649 |
+
"model.layers.7.self_attn.k_proj.weight": "model.safetensors",
|
| 650 |
+
"model.layers.7.self_attn.o_proj.biases": "model.safetensors",
|
| 651 |
+
"model.layers.7.self_attn.o_proj.scales": "model.safetensors",
|
| 652 |
+
"model.layers.7.self_attn.o_proj.weight": "model.safetensors",
|
| 653 |
+
"model.layers.7.self_attn.q_norm.weight": "model.safetensors",
|
| 654 |
+
"model.layers.7.self_attn.q_proj.biases": "model.safetensors",
|
| 655 |
+
"model.layers.7.self_attn.q_proj.scales": "model.safetensors",
|
| 656 |
+
"model.layers.7.self_attn.q_proj.weight": "model.safetensors",
|
| 657 |
+
"model.layers.7.self_attn.v_proj.biases": "model.safetensors",
|
| 658 |
+
"model.layers.7.self_attn.v_proj.scales": "model.safetensors",
|
| 659 |
+
"model.layers.7.self_attn.v_proj.weight": "model.safetensors",
|
| 660 |
+
"model.layers.8.input_layernorm.weight": "model.safetensors",
|
| 661 |
+
"model.layers.8.mlp.down_proj.biases": "model.safetensors",
|
| 662 |
+
"model.layers.8.mlp.down_proj.scales": "model.safetensors",
|
| 663 |
+
"model.layers.8.mlp.down_proj.weight": "model.safetensors",
|
| 664 |
+
"model.layers.8.mlp.gate_proj.biases": "model.safetensors",
|
| 665 |
+
"model.layers.8.mlp.gate_proj.scales": "model.safetensors",
|
| 666 |
+
"model.layers.8.mlp.gate_proj.weight": "model.safetensors",
|
| 667 |
+
"model.layers.8.mlp.up_proj.biases": "model.safetensors",
|
| 668 |
+
"model.layers.8.mlp.up_proj.scales": "model.safetensors",
|
| 669 |
+
"model.layers.8.mlp.up_proj.weight": "model.safetensors",
|
| 670 |
+
"model.layers.8.post_attention_layernorm.weight": "model.safetensors",
|
| 671 |
+
"model.layers.8.self_attn.k_norm.weight": "model.safetensors",
|
| 672 |
+
"model.layers.8.self_attn.k_proj.biases": "model.safetensors",
|
| 673 |
+
"model.layers.8.self_attn.k_proj.scales": "model.safetensors",
|
| 674 |
+
"model.layers.8.self_attn.k_proj.weight": "model.safetensors",
|
| 675 |
+
"model.layers.8.self_attn.o_proj.biases": "model.safetensors",
|
| 676 |
+
"model.layers.8.self_attn.o_proj.scales": "model.safetensors",
|
| 677 |
+
"model.layers.8.self_attn.o_proj.weight": "model.safetensors",
|
| 678 |
+
"model.layers.8.self_attn.q_norm.weight": "model.safetensors",
|
| 679 |
+
"model.layers.8.self_attn.q_proj.biases": "model.safetensors",
|
| 680 |
+
"model.layers.8.self_attn.q_proj.scales": "model.safetensors",
|
| 681 |
+
"model.layers.8.self_attn.q_proj.weight": "model.safetensors",
|
| 682 |
+
"model.layers.8.self_attn.v_proj.biases": "model.safetensors",
|
| 683 |
+
"model.layers.8.self_attn.v_proj.scales": "model.safetensors",
|
| 684 |
+
"model.layers.8.self_attn.v_proj.weight": "model.safetensors",
|
| 685 |
+
"model.layers.9.input_layernorm.weight": "model.safetensors",
|
| 686 |
+
"model.layers.9.mlp.down_proj.biases": "model.safetensors",
|
| 687 |
+
"model.layers.9.mlp.down_proj.scales": "model.safetensors",
|
| 688 |
+
"model.layers.9.mlp.down_proj.weight": "model.safetensors",
|
| 689 |
+
"model.layers.9.mlp.gate_proj.biases": "model.safetensors",
|
| 690 |
+
"model.layers.9.mlp.gate_proj.scales": "model.safetensors",
|
| 691 |
+
"model.layers.9.mlp.gate_proj.weight": "model.safetensors",
|
| 692 |
+
"model.layers.9.mlp.up_proj.biases": "model.safetensors",
|
| 693 |
+
"model.layers.9.mlp.up_proj.scales": "model.safetensors",
|
| 694 |
+
"model.layers.9.mlp.up_proj.weight": "model.safetensors",
|
| 695 |
+
"model.layers.9.post_attention_layernorm.weight": "model.safetensors",
|
| 696 |
+
"model.layers.9.self_attn.k_norm.weight": "model.safetensors",
|
| 697 |
+
"model.layers.9.self_attn.k_proj.biases": "model.safetensors",
|
| 698 |
+
"model.layers.9.self_attn.k_proj.scales": "model.safetensors",
|
| 699 |
+
"model.layers.9.self_attn.k_proj.weight": "model.safetensors",
|
| 700 |
+
"model.layers.9.self_attn.o_proj.biases": "model.safetensors",
|
| 701 |
+
"model.layers.9.self_attn.o_proj.scales": "model.safetensors",
|
| 702 |
+
"model.layers.9.self_attn.o_proj.weight": "model.safetensors",
|
| 703 |
+
"model.layers.9.self_attn.q_norm.weight": "model.safetensors",
|
| 704 |
+
"model.layers.9.self_attn.q_proj.biases": "model.safetensors",
|
| 705 |
+
"model.layers.9.self_attn.q_proj.scales": "model.safetensors",
|
| 706 |
+
"model.layers.9.self_attn.q_proj.weight": "model.safetensors",
|
| 707 |
+
"model.layers.9.self_attn.v_proj.biases": "model.safetensors",
|
| 708 |
+
"model.layers.9.self_attn.v_proj.scales": "model.safetensors",
|
| 709 |
+
"model.layers.9.self_attn.v_proj.weight": "model.safetensors",
|
| 710 |
+
"model.norm.weight": "model.safetensors"
|
| 711 |
+
}
|
| 712 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
| 3 |
+
size 11422650
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": null,
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"extra_special_tokens": [
|
| 9 |
+
"<|im_start|>",
|
| 10 |
+
"<|im_end|>",
|
| 11 |
+
"<|object_ref_start|>",
|
| 12 |
+
"<|object_ref_end|>",
|
| 13 |
+
"<|box_start|>",
|
| 14 |
+
"<|box_end|>",
|
| 15 |
+
"<|quad_start|>",
|
| 16 |
+
"<|quad_end|>",
|
| 17 |
+
"<|vision_start|>",
|
| 18 |
+
"<|vision_end|>",
|
| 19 |
+
"<|vision_pad|>",
|
| 20 |
+
"<|image_pad|>",
|
| 21 |
+
"<|video_pad|>"
|
| 22 |
+
],
|
| 23 |
+
"is_local": true,
|
| 24 |
+
"local_files_only": false,
|
| 25 |
+
"model_max_length": 131072,
|
| 26 |
+
"pad_token": "<|endoftext|>",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"tool_parser_type": "json_tools",
|
| 30 |
+
"unk_token": null
|
| 31 |
+
}
|