MyeongHoJeong commited on
Commit
4581153
·
verified ·
1 Parent(s): 7404fef

Standard One 3B SH v1

Browse files
.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
+ tekken.json filter=lfs diff=lfs merge=lfs -text
37
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
LICENSE ADDED
@@ -0,0 +1,201 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Apache License
2
+ Version 2.0, January 2004
3
+ http://www.apache.org/licenses/
4
+
5
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
6
+
7
+ 1. Definitions.
8
+
9
+ "License" shall mean the terms and conditions for use, reproduction,
10
+ and distribution as defined by Sections 1 through 9 of this document.
11
+
12
+ "Licensor" shall mean the copyright owner or entity authorized by
13
+ the copyright owner that is granting the License.
14
+
15
+ "Legal Entity" shall mean the union of the acting entity and all
16
+ other entities that control, are controlled by, or are under common
17
+ control with that entity. For the purposes of this definition,
18
+ "control" means (i) the power, direct or indirect, to cause the
19
+ direction or management of such entity, whether by contract or
20
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
21
+ outstanding shares, or (iii) beneficial ownership of such entity.
22
+
23
+ "You" (or "Your") shall mean an individual or Legal Entity
24
+ exercising permissions granted by this License.
25
+
26
+ "Source" form shall mean the preferred form for making modifications,
27
+ including but not limited to software source code, documentation
28
+ source, and configuration files.
29
+
30
+ "Object" form shall mean any form resulting from mechanical
31
+ transformation or translation of a Source form, including but
32
+ not limited to compiled object code, generated documentation,
33
+ and conversions to other media types.
34
+
35
+ "Work" shall mean the work of authorship, whether in Source or
36
+ Object form, made available under the License, as indicated by a
37
+ copyright notice that is included in or attached to the work
38
+ (an example is provided in the Appendix below).
39
+
40
+ "Derivative Works" shall mean any work, whether in Source or Object
41
+ form, that is based on (or derived from) the Work and for which the
42
+ editorial revisions, annotations, elaborations, or other modifications
43
+ represent, as a whole, an original work of authorship. For the purposes
44
+ of this License, Derivative Works shall not include works that remain
45
+ separable from, or merely link (or bind by name) to the interfaces of,
46
+ the Work and Derivative Works thereof.
47
+
48
+ "Contribution" shall mean any work of authorship, including
49
+ the original version of the Work and any modifications or additions
50
+ to that Work or Derivative Works thereof, that is intentionally
51
+ submitted to Licensor for inclusion in the Work by the copyright owner
52
+ or by an individual or Legal Entity authorized to submit on behalf of
53
+ the copyright owner. For the purposes of this definition, "submitted"
54
+ means any form of electronic, verbal, or written communication sent
55
+ to the Licensor or its representatives, including but not limited to
56
+ communication on electronic mailing lists, source code control systems,
57
+ and issue tracking systems that are managed by, or on behalf of, the
58
+ Licensor for the purpose of discussing and improving the Work, but
59
+ excluding communication that is conspicuously marked or otherwise
60
+ designated in writing by the copyright owner as "Not a Contribution."
61
+
62
+ "Contributor" shall mean Licensor and any individual or Legal Entity
63
+ on behalf of whom a Contribution has been received by Licensor and
64
+ subsequently incorporated within the Work.
65
+
66
+ 2. Grant of Copyright License. Subject to the terms and conditions of
67
+ this License, each Contributor hereby grants to You a perpetual,
68
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
69
+ copyright license to reproduce, prepare Derivative Works of,
70
+ publicly display, publicly perform, sublicense, and distribute the
71
+ Work and such Derivative Works in Source or Object form.
72
+
73
+ 3. Grant of Patent License. Subject to the terms and conditions of
74
+ this License, each Contributor hereby grants to You a perpetual,
75
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
76
+ (except as stated in this section) patent license to make, have made,
77
+ use, offer to sell, sell, import, and otherwise transfer the Work,
78
+ where such license applies only to those patent claims licensable
79
+ by such Contributor that are necessarily infringed by their
80
+ Contribution(s) alone or by combination of their Contribution(s)
81
+ with the Work to which such Contribution(s) was submitted. If You
82
+ institute patent litigation against any entity (including a
83
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
84
+ or a Contribution incorporated within the Work constitutes direct
85
+ or contributory patent infringement, then any patent licenses
86
+ granted to You under this License for that Work shall terminate
87
+ as of the date such litigation is filed.
88
+
89
+ 4. Redistribution. You may reproduce and distribute copies of the
90
+ Work or Derivative Works thereof in any medium, with or without
91
+ modifications, and in Source or Object form, provided that You
92
+ meet the following conditions:
93
+
94
+ (a) You must give any other recipients of the Work or
95
+ Derivative Works a copy of this License; and
96
+
97
+ (b) You must cause any modified files to carry prominent notices
98
+ stating that You changed the files; and
99
+
100
+ (c) You must retain, in the Source form of any Derivative Works
101
+ that You distribute, all copyright, patent, trademark, and
102
+ attribution notices from the Source form of the Work,
103
+ excluding those notices that do not pertain to any part of
104
+ the Derivative Works; and
105
+
106
+ (d) If the Work includes a "NOTICE" text file as part of its
107
+ distribution, then any Derivative Works that You distribute must
108
+ include a readable copy of the attribution notices contained
109
+ within such NOTICE file, excluding those notices that do not
110
+ pertain to any part of the Derivative Works, in at least one
111
+ of the following places: within a NOTICE text file distributed
112
+ as part of the Derivative Works; within the Source form or
113
+ documentation, if provided along with the Derivative Works; or,
114
+ within a display generated by the Derivative Works, if and
115
+ wherever such third-party notices normally appear. The contents
116
+ of the NOTICE file are for informational purposes only and
117
+ do not modify the License. You may add Your own attribution
118
+ notices within Derivative Works that You distribute, alongside
119
+ or as an addendum to the NOTICE text from the Work, provided
120
+ that such additional attribution notices cannot be construed
121
+ as modifying the License.
122
+
123
+ You may add Your own copyright statement to Your modifications and
124
+ may provide additional or different license terms and conditions
125
+ for use, reproduction, or distribution of Your modifications, or
126
+ for any such Derivative Works as a whole, provided Your use,
127
+ reproduction, and distribution of the Work otherwise complies with
128
+ the conditions stated in this License.
129
+
130
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
131
+ any Contribution intentionally submitted for inclusion in the Work
132
+ by You to the Licensor shall be under the terms and conditions of
133
+ this License, without any additional terms or conditions.
134
+ Notwithstanding the above, nothing herein shall supersede or modify
135
+ the terms of any separate license agreement you may have executed
136
+ with Licensor regarding such Contributions.
137
+
138
+ 6. Trademarks. This License does not grant permission to use the trade
139
+ names, trademarks, service marks, or product names of the Licensor,
140
+ except as required for reasonable and customary use in describing the
141
+ origin of the Work and reproducing the content of the NOTICE file.
142
+
143
+ 7. Disclaimer of Warranty. Unless required by applicable law or
144
+ agreed to in writing, Licensor provides the Work (and each
145
+ Contributor provides its Contributions) on an "AS IS" BASIS,
146
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
147
+ implied, including, without limitation, any warranties or conditions
148
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
149
+ PARTICULAR PURPOSE. You are solely responsible for determining the
150
+ appropriateness of using or redistributing the Work and assume any
151
+ risks associated with Your exercise of permissions under this License.
152
+
153
+ 8. Limitation of Liability. In no event and under no legal theory,
154
+ whether in tort (including negligence), contract, or otherwise,
155
+ unless required by applicable law (such as deliberate and grossly
156
+ negligent acts) or agreed to in writing, shall any Contributor be
157
+ liable to You for damages, including any direct, indirect, special,
158
+ incidental, or consequential damages of any character arising as a
159
+ result of this License or out of the use or inability to use the
160
+ Work (including but not limited to damages for loss of goodwill,
161
+ work stoppage, computer failure or malfunction, or any and all
162
+ other commercial damages or losses), even if such Contributor
163
+ has been advised of the possibility of such damages.
164
+
165
+ 9. Accepting Warranty or Additional Liability. While redistributing
166
+ the Work or Derivative Works thereof, You may choose to offer,
167
+ and charge a fee for, acceptance of support, warranty, indemnity,
168
+ or other liability obligations and/or rights consistent with this
169
+ License. However, in accepting such obligations, You may act only
170
+ on Your own behalf and on Your sole responsibility, not on behalf
171
+ of any other Contributor, and only if You agree to indemnify,
172
+ defend, and hold each Contributor harmless for any liability
173
+ incurred by, or claims asserted against, such Contributor by reason
174
+ of your accepting any such warranty or additional liability.
175
+
176
+ END OF TERMS AND CONDITIONS
177
+
178
+ APPENDIX: How to apply the Apache License to your work.
179
+
180
+ To apply the Apache License to your work, attach the following
181
+ boilerplate notice, with the fields enclosed by brackets "[]"
182
+ replaced with your own identifying information. (Don't include
183
+ the brackets!) The text should be enclosed in the appropriate
184
+ comment syntax for the file format. We also recommend that a
185
+ file or class name and description of purpose be included on the
186
+ same "printed page" as the copyright notice for easier
187
+ identification within third-party archives.
188
+
189
+ Copyright 2023-2024 SGLang Team
190
+
191
+ Licensed under the Apache License, Version 2.0 (the "License");
192
+ you may not use this file except in compliance with the License.
193
+ You may obtain a copy of the License at
194
+
195
+ http://www.apache.org/licenses/LICENSE-2.0
196
+
197
+ Unless required by applicable law or agreed to in writing, software
198
+ distributed under the License is distributed on an "AS IS" BASIS,
199
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
200
+ See the License for the specific language governing permissions and
201
+ limitations under the License.
README.md ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: StandardThinking/StandardOne-3B-SH
3
+ base_model_relation: quantized
4
+ library_name: transformers
5
+ license: apache-2.0
6
+ pipeline_tag: text-classification
7
+ tags:
8
+ - mistral3
9
+ - decision-model
10
+ - schema-head
11
+ - fp8
12
+ - compressed-tensors
13
+ ---
14
+
15
+ # Standard One 3B SH — FP8
16
+
17
+ **Version:** v1 (2026-10-09)
18
+
19
+ An FP8 (compressed-tensors, float8_e4m3 weights, dynamic per-token activations) quantization of the merged backbone
20
+ of [Standard One 3B SH](https://huggingface.co/StandardThinking/StandardOne-3B-SH) v1, with the same schema head.
21
+ The language-model linear projections (q/k/v/o, gate/up/down; 182 tensors) are quantized per output channel
22
+ (scale = absmax / 448); the vision tower, multi-modal projector, embeddings, norms and lm_head stay in BF16. This is
23
+ the same data-free recipe and `quantization_config` as
24
+ [StandardOne-3B-FP8](https://huggingface.co/StandardThinking/StandardOne-3B-FP8). The head (`head/`) is float32 and
25
+ identical to the BF16 release.
26
+
27
+ | | |
28
+ |---|---|
29
+ | Weights | 4.7 GB (BF16 release: 7.7 GB) |
30
+ | Head | `head/schema_head.safetensors`, float32, taps: final layer and layer 19 |
31
+ | Prompt format | chat template with no default system message (see the [main card](https://huggingface.co/StandardThinking/StandardOne-3B-SH#prompt-format)) |
32
+
33
+ ## Validation
34
+
35
+ Both precisions were served the same way by our SGLang-based decision engine (schema-head readout, bfloat16
36
+ activations, one option order), measured 2026-10-09. Accuracy is the most probable answer.
37
+
38
+ | Suite | BF16 (StandardOne-3B-SH v1) | FP8 (this repository) | Change (points) |
39
+ |---|---:|---:|---:|
40
+ | many-option questions, 53–151 options (18,000) | 92.23 % | 92.15 % | −0.08 |
41
+ | the same question set, at most 26 options (750) | 94.67 % | 94.80 % | +0.13 |
42
+ | long-document questions (150) | 82.00 % | 83.33 % | +1.33 |
43
+ | held-out decision set (600) | 83.17 % | 82.50 % | −0.67 |
44
+ | hard proxy (600) | 40.17 % | 39.50 % | −0.67 |
45
+ | realistic transfer set (600) | 88.67 % | 88.00 % | −0.67 |
46
+ | WinoGrande dev (2,010) | 92.69 % | 92.39 % | −0.30 |
47
+ | policy-judgement repair (1,200) | 68.08 % | 68.00 % | −0.08 |
48
+ | JevBench public easy (48) | 100.00 % | 100.00 % | +0.00 |
49
+ | JevBench public standard (72) | 97.22 % | 97.22 % | +0.00 |
50
+ | JevBench public hard (111) | 54.05 % | 54.95 % | +0.90 |
51
+
52
+ Across 4,791 validation questions, FP8 and BF16 gave the same answer for 97.70 %.
53
+
54
+ ## Notes
55
+
56
+ - Engines that load compressed-tensors FP8 (SGLang, vLLM) read this checkpoint. On ROCm with the native per-token
57
+ activation-quantization fallback, SGLang 0.5.20 needs a small fix when the engine pads the token dimension
58
+ (the per-token scale buffer is padded but the fallback writes unpadded rows); without it the server fails at
59
+ warmup. The validation above used that fix.
60
+ - The PyTorch reference server in StandardOne-3B-SH is written for the BF16 checkpoint; use the BF16 release with it.
61
+
62
+ ## License
63
+
64
+ Apache-2.0, the same terms as [StandardOne-3B-SH](https://huggingface.co/StandardThinking/StandardOne-3B-SH),
65
+ built on [Standard One 3B](https://huggingface.co/StandardThinking/StandardOne-3B) and
66
+ [Ministral 3 3B](https://huggingface.co/mistralai/Ministral-3-3B-Instruct-2512) (Apache-2.0).
SHA256SUMS ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 1495e1e757ef4d0925a2350563cf5754bb23c51701a8ec4fb3c5cdcbedae6747 LICENSE
2
+ 73bfdeb21a28e74b52dcd9b0f835215d7a3e92d12e54f1461721590cd7a8770a README.md
3
+ 331b249682cd52226c50e533f59825184997f3b31f9a62b2c3fea940db6999c5 SYSTEM_PROMPT.txt
4
+ e62f05da25555a4c109d0f413eb673d328fcc40f7c82537c61631611f6954372 chat_template.jinja
5
+ 26615a64d8e6b0955811ae6bc4bb1200dc2c91df4a10474d33968f0fe70b82a4 config.json
6
+ e0923390059f84a9180b00e5501778acc45ea9856cd7f2fd68208b360927c677 generation_config.json
7
+ 7c7b61dd2b8bee2557b1384f08af37c0992963339c418ef467fa74fc3b3ee962 head/schema_head.safetensors
8
+ 8b5004712e6205726f97a08f2a7edd11bcdc418ca9cca24698aa707d5a254373 head/schema_head_config.json
9
+ 32c9aa6a15284a7491546f5d69fe8a12af4bab27f7592ca596c8eec807f95de6 model-00001-of-00002.safetensors
10
+ d253eb9a626fde1efa54ec898bd4eb20afb1774c814269582411197e412275aa model-00002-of-00002.safetensors
11
+ 9d7ff6eac886544535cd4d956c88b64726ceaac4ef3ab9f46f59520b9061ad8a model.safetensors.index.json
12
+ d15587c57574d2a92c66ff2552c6db5173d4c440ad47f7454dc51bad0ecaeb59 params.json
13
+ ece2373c2ae391bce18785a1810543bc6173a1f28b6767cffab2c35dbea5f002 processor_config.json
14
+ 32fb95d7d1f6b64b205a327074072dc777b98acab8beaaa031806474bf406ccb recipe.yaml
15
+ 9285e4edaeb41a43d6fdfa9a2dd70fe739e09f15f3f7f976dc9932d590a919ae release-manifest.json
16
+ 0a5c981e8c5c6f8886ee007a6d4543a0be6b221cb9ca32a8709384a4c6fc8cbb special_tokens_map.json
17
+ 600bb27946565481ecf51ba8aee252e49b9a68507866080ac9c30185bb312843 tekken.json
18
+ d5f6046775b112f0e2d456ee9dba450684ab964fe5c4e231599bdc6773028135 tokenizer.json
19
+ f59f7294e4f26383d0ea93840fe21cf197784be0842a8301a0343e8c34ed0d6d tokenizer_config.json
SYSTEM_PROMPT.txt ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ You are Ministral-3-3B-Instruct-2512, a Large Language Model (LLM) created by Mistral AI, a French startup headquartered in Paris.
2
+ You power an AI assistant called Le Chat.
3
+ Your knowledge base was last updated on 2023-10-01.
4
+ The current date is {today}.
5
+
6
+ When you're not sure about some information or when the user's request requires up-to-date or specific data, you must use the available tools to fetch the information. Do not hesitate to use tools whenever they can provide a more accurate or complete response. If no relevant tools are available, then clearly state that you don't have the information and avoid making up anything.
7
+ If the user's question is not clear, ambiguous, or does not provide enough context for you to accurately answer the question, you do not try to answer it right away and you rather ask the user to clarify their request (e.g. "What are some good restaurants around me?" => "Where are you?" or "When is the next flight to Tokyo" => "Where do you travel from?").
8
+ You are always very attentive to dates, in particular you try to resolve dates (e.g. "yesterday" is {yesterday}) and when asked about information at specific dates, you discard information that is at another date.
9
+ You follow these instructions in all languages, and always respond to the user in the language they use or request.
10
+ Next sections describe the capabilities that you have.
11
+
12
+ # WEB BROWSING INSTRUCTIONS
13
+
14
+ You cannot perform any web search or access internet to open URLs, links etc. If it seems like the user is expecting you to do so, you clarify the situation and ask the user to copy paste the text directly in the chat.
15
+
16
+ # MULTI-MODAL INSTRUCTIONS
17
+
18
+ You have the ability to read images, but you cannot generate images. You also cannot transcribe audio files or videos.
19
+ You cannot read nor transcribe audio files or videos.
20
+
21
+ # TOOL CALLING INSTRUCTIONS
22
+
23
+ You may have access to tools that you can use to fetch information or perform actions. You must use these tools in the following situations:
24
+
25
+ 1. When the request requires up-to-date information.
26
+ 2. When the request requires specific data that you do not have in your knowledge base.
27
+ 3. When the request involves actions that you cannot perform without tools.
28
+
29
+ Always prioritize using tools to provide the most accurate and helpful response. If tools are not available, inform the user that you cannot perform the requested action at the moment.
chat_template.jinja ADDED
@@ -0,0 +1,189 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {#- Default system message if no system prompt is passed. #}
2
+ {%- set default_system_message = '' %}
3
+
4
+ {#- Begin of sequence token. #}
5
+ {{- bos_token }}
6
+
7
+
8
+ {#- Handle system prompt if it exists. #}
9
+ {%- set loop_messages = messages %}
10
+ {%- if messages[0]['role'] != 'system' and default_system_message != '' %}
11
+ {{- '[SYSTEM_PROMPT]' + default_system_message + '[/SYSTEM_PROMPT]' }}
12
+ {%- endif %}
13
+
14
+
15
+ {#- Tools definition #}
16
+ {%- set available_tools = '' %}
17
+ {%- set has_tools = false %}
18
+ {%- if tools is defined and tools is not none and tools|length > 0 %}
19
+ {%- set has_tools = true %}
20
+ {%- set available_tools = '[AVAILABLE_TOOLS]' + (tools| tojson) + '[/AVAILABLE_TOOLS]' %}
21
+ {%- endif %}
22
+
23
+ {#- Macros #}
24
+ {%- macro render_content(content, context_name, supported_types_desc, support_images) -%}
25
+ {%- if content is string -%}
26
+ {{- content -}}
27
+ {%- elif content -%}
28
+ {%- for block in content -%}
29
+ {%- if block['type'] == 'text' -%}
30
+ {{- block['text'] -}}
31
+ {%- elif support_images and block['type'] in ['image', 'image_url'] -%}
32
+ {{- '[IMG]' -}}
33
+ {%- else -%}
34
+ {{- raise_exception('Only ' + supported_types_desc + ' chunks are supported in ' + context_name + '.') -}}
35
+ {%- endif -%}
36
+ {%- endfor -%}
37
+ {%- else -%}
38
+ {{- raise_exception(context_name + ' must have non-empty content.') -}}
39
+ {%- endif -%}
40
+ {%- endmacro -%}
41
+
42
+ {#- Aggregate consecutive messages with the same role except system and tool. #}
43
+ {#- A sentinel message is appended so the last group gets flushed inside the loop. #}
44
+ {%- set ns_agg = namespace(messages=[], current_group=[], current_role=none) %}
45
+ {%- for message in loop_messages + [{'role': '__sentinel__'}] %}
46
+ {%- if message['role'] != ns_agg.current_role or message['role'] == 'system' or message['role'] == 'tool' %}
47
+ {%- if ns_agg.current_role == 'tool' %}
48
+ {%- set ns_agg.messages = ns_agg.messages + ns_agg.current_group %}
49
+ {%- elif ns_agg.current_role is not none %}
50
+ {%- set ns_c = namespace(text_parts=[], chunks=[], has_non_text=false, tool_calls=[]) %}
51
+ {%- for msg in ns_agg.current_group %}
52
+ {%- if msg['content'] is string %}
53
+ {%- set ns_c.text_parts = ns_c.text_parts + [msg['content']] %}
54
+ {%- elif msg['content'] is not none %}
55
+ {%- for block in msg['content'] %}
56
+ {%- if block['type'] == 'text' %}
57
+ {%- set ns_c.text_parts = ns_c.text_parts + [block['text']] %}
58
+ {%- else %}
59
+ {%- if ns_c.text_parts | length > 0 %}
60
+ {%- set ns_c.chunks = ns_c.chunks + [{'type': 'text', 'text': ns_c.text_parts | join('\n\n')}] %}
61
+ {%- set ns_c.text_parts = [] %}
62
+ {%- endif %}
63
+ {%- set ns_c.chunks = ns_c.chunks + [block] %}
64
+ {%- set ns_c.has_non_text = true %}
65
+ {%- endif %}
66
+ {%- endfor %}
67
+ {%- endif %}
68
+ {%- if msg['tool_calls'] is defined and msg['tool_calls'] is not none %}
69
+ {%- set ns_c.tool_calls = ns_c.tool_calls + msg['tool_calls'] | list %}
70
+ {%- endif %}
71
+ {%- endfor %}
72
+ {%- if ns_c.has_non_text %}
73
+ {%- if ns_c.text_parts | length > 0 %}
74
+ {%- set ns_c.chunks = ns_c.chunks + [{'type': 'text', 'text': ns_c.text_parts | join('\n\n')}] %}
75
+ {%- endif %}
76
+ {%- set merged_content = ns_c.chunks %}
77
+ {%- else %}
78
+ {%- set merged_content = ns_c.text_parts | join('\n\n') %}
79
+ {%- endif %}
80
+ {%- if ns_c.tool_calls | length > 0 %}
81
+ {%- set ns_agg.messages = ns_agg.messages + [{'role': ns_agg.current_role, 'content': merged_content, 'tool_calls': ns_c.tool_calls}] %}
82
+ {%- else %}
83
+ {%- set ns_agg.messages = ns_agg.messages + [{'role': ns_agg.current_role, 'content': merged_content}] %}
84
+ {%- endif %}
85
+ {%- endif %}
86
+ {%- if message['role'] != '__sentinel__' %}
87
+ {%- set ns_agg.current_group = [message] %}
88
+ {%- set ns_agg.current_role = message['role'] %}
89
+ {%- endif %}
90
+ {%- else %}
91
+ {%- set ns_agg.current_group = ns_agg.current_group + [message] %}
92
+ {%- endif %}
93
+ {%- endfor %}
94
+ {%- set loop_messages = ns_agg.messages %}
95
+
96
+ {#- Validates message ordering. #}
97
+ {%- set ns = namespace(available_tools_emitted=false) %}
98
+ {%- if loop_messages | length > 0 and loop_messages[0]['role'] not in ['user', 'system'] %}
99
+ {{- raise_exception('Conversation must start with a user or system message, got ' + loop_messages[0]['role'] + '.') }}
100
+ {%- endif %}
101
+ {%- set ns_order = namespace(previous_role=none) %}
102
+ {%- for message in loop_messages %}
103
+ {%- set current_role = message['role'] %}
104
+ {%- if ns_order.previous_role is not none %}
105
+ {%- if ns_order.previous_role == 'system' %}
106
+ {%- if current_role not in ['user', 'assistant', 'system'] %}
107
+ {{- raise_exception('Unexpected role \'' + current_role + '\' after role \'' + ns_order.previous_role + '\'') }}
108
+ {%- endif %}
109
+ {%- elif ns_order.previous_role == 'user' %}
110
+ {%- if current_role not in ['assistant', 'system', 'user'] %}
111
+ {{- raise_exception('Unexpected role \'' + current_role + '\' after role \'' + ns_order.previous_role + '\'') }}
112
+ {%- endif %}
113
+ {%- elif ns_order.previous_role == 'assistant' %}
114
+ {%- if current_role not in ['assistant', 'user', 'tool'] %}
115
+ {{- raise_exception('Unexpected role \'' + current_role + '\' after role \'' + ns_order.previous_role + '\'') }}
116
+ {%- endif %}
117
+ {%- elif ns_order.previous_role == 'tool' %}
118
+ {%- if current_role not in ['assistant', 'tool', 'user'] %}
119
+ {{- raise_exception('Unexpected role \'' + current_role + '\' after role \'' + ns_order.previous_role + '\'') }}
120
+ {%- endif %}
121
+ {%- endif %}
122
+ {%- endif %}
123
+ {%- set ns_order.previous_role = current_role %}
124
+ {%- endfor %}
125
+
126
+ {#- Handle conversation messages. #}
127
+ {%- for message in loop_messages %}
128
+ {#- User messages supports text, image and image_url content. #}
129
+ {%- if message['role'] == 'user' %}
130
+ {%- if not ns.available_tools_emitted %}
131
+ {{- available_tools }}
132
+ {%- set ns.available_tools_emitted = true %}
133
+ {%- endif %}
134
+ {%- if message['content'] is not string and message['content'] %}
135
+ {#- When content has exactly one image and one text block, put image first. #}
136
+ {%- if message['content'] | length == 2 and message['content'][0]['type'] == 'text' and message['content'][1]['type'] in ['image', 'image_url'] %}
137
+ {%- set blocks = [message['content'][1], message['content'][0]] %}
138
+ {%- else %}
139
+ {%- set blocks = message['content'] %}
140
+ {%- endif %}
141
+ {%- set user_content = blocks %}
142
+ {%- else %}
143
+ {%- set user_content = message['content'] %}
144
+ {%- endif %}
145
+ {{- '[INST]' -}}
146
+ {{- render_content(user_content, 'user message content', supported_types_desc='text, image and image_url', support_images=true) -}}
147
+ {{- '[/INST]' }}
148
+
149
+ {#- Assistant messages supports text content. #}
150
+ {%- elif message['role'] == 'assistant' %}
151
+ {%- if (message['content'] is none or message['content'] == '' or message['content']|length == 0) and (message['tool_calls'] is not defined or message['tool_calls'] is none or message['tool_calls']|length == 0) %}
152
+ {{- raise_exception('Assistant message must have a string or a list of chunks in content or a list of tool calls.') }}
153
+ {%- endif %}
154
+
155
+ {%- if message['content'] %}
156
+ {{- render_content(message['content'], 'assistant message contents', supported_types_desc='text', support_images=false) -}}
157
+ {%- endif %}
158
+
159
+ {%- if message['tool_calls'] is defined and message['tool_calls'] is not none and message['tool_calls']|length > 0 %}
160
+ {%- for tool in message['tool_calls'] %}
161
+ {{- '[TOOL_CALLS]' }}
162
+ {%- set name = tool['function']['name'] %}
163
+ {%- set arguments = tool['function']['arguments'] %}
164
+ {%- if arguments is not string %}
165
+ {%- set arguments = arguments|tojson|safe %}
166
+ {%- elif arguments == '' %}
167
+ {%- set arguments = '{}' %}
168
+ {%- endif %}
169
+ {{- name + '[ARGS]' + arguments }}
170
+ {%- endfor %}
171
+ {%- endif %}
172
+
173
+ {{- eos_token }}
174
+
175
+ {#- Tool messages only supports text content. #}
176
+ {%- elif message['role'] == 'tool' %}
177
+ {{- '[TOOL_RESULTS]' + message['content']|string + '[/TOOL_RESULTS]' }}
178
+
179
+ {#- System messages. #}
180
+ {%- elif message['role'] == 'system' %}
181
+ {{- '[SYSTEM_PROMPT]' -}}
182
+ {{- render_content(message['content'], 'system message contents', supported_types_desc='text', support_images=false) -}}
183
+ {{- '[/SYSTEM_PROMPT]' -}}
184
+
185
+ {#- Raise exception for unsupported roles. #}
186
+ {%- else %}
187
+ {{- raise_exception('Only user, assistant, system and tool roles are supported, got ' + message['role'] + '.') }}
188
+ {%- endif %}
189
+ {%- endfor %}
config.json ADDED
@@ -0,0 +1,288 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Mistral3ForConditionalGeneration"
4
+ ],
5
+ "dtype": "bfloat16",
6
+ "image_token_index": 10,
7
+ "model_type": "mistral3",
8
+ "multimodal_projector_bias": false,
9
+ "projector_hidden_act": "gelu",
10
+ "spatial_merge_size": 2,
11
+ "text_config": {
12
+ "attention_dropout": 0.0,
13
+ "bos_token_id": 1,
14
+ "dtype": "float32",
15
+ "eos_token_id": 2,
16
+ "head_dim": 128,
17
+ "hidden_act": "silu",
18
+ "hidden_size": 3072,
19
+ "initializer_range": 0.02,
20
+ "intermediate_size": 9216,
21
+ "max_position_embeddings": 262144,
22
+ "model_type": "ministral3",
23
+ "num_attention_heads": 32,
24
+ "num_hidden_layers": 26,
25
+ "num_key_value_heads": 8,
26
+ "pad_token_id": 11,
27
+ "rms_norm_eps": 1e-05,
28
+ "rope_parameters": {
29
+ "beta_fast": 32.0,
30
+ "beta_slow": 1.0,
31
+ "factor": 16.0,
32
+ "llama_4_scaling_beta": 0.1,
33
+ "mscale": 1.0,
34
+ "mscale_all_dim": 1.0,
35
+ "original_max_position_embeddings": 16384,
36
+ "rope_theta": 1000000.0,
37
+ "rope_type": "yarn",
38
+ "type": "yarn"
39
+ },
40
+ "sliding_window": null,
41
+ "tie_word_embeddings": true,
42
+ "use_cache": true,
43
+ "vocab_size": 131072
44
+ },
45
+ "tie_word_embeddings": true,
46
+ "transformers_version": "5.12.1",
47
+ "vision_config": {
48
+ "attention_dropout": 0.0,
49
+ "dtype": "float32",
50
+ "head_dim": 64,
51
+ "hidden_act": "silu",
52
+ "hidden_size": 1024,
53
+ "image_size": 1540,
54
+ "initializer_range": 0.02,
55
+ "intermediate_size": 4096,
56
+ "model_type": "pixtral",
57
+ "num_attention_heads": 16,
58
+ "num_channels": 3,
59
+ "num_hidden_layers": 24,
60
+ "patch_size": 14,
61
+ "rope_parameters": {
62
+ "rope_theta": 10000.0,
63
+ "rope_type": "default"
64
+ }
65
+ },
66
+ "vision_feature_layer": -1,
67
+ "quantization_config": {
68
+ "config_groups": {
69
+ "group_0": {
70
+ "format": "float-quantized",
71
+ "input_activations": {
72
+ "actorder": null,
73
+ "block_structure": null,
74
+ "dynamic": true,
75
+ "group_size": null,
76
+ "num_bits": 8,
77
+ "observer": null,
78
+ "observer_kwargs": {},
79
+ "scale_dtype": null,
80
+ "strategy": "token",
81
+ "symmetric": true,
82
+ "type": "float",
83
+ "zp_dtype": null
84
+ },
85
+ "output_activations": null,
86
+ "targets": [
87
+ "Linear"
88
+ ],
89
+ "weights": {
90
+ "actorder": null,
91
+ "block_structure": null,
92
+ "dynamic": false,
93
+ "group_size": null,
94
+ "num_bits": 8,
95
+ "observer": "memoryless_minmax",
96
+ "observer_kwargs": {},
97
+ "scale_dtype": null,
98
+ "strategy": "channel",
99
+ "symmetric": true,
100
+ "type": "float",
101
+ "zp_dtype": null
102
+ }
103
+ }
104
+ },
105
+ "format": "float-quantized",
106
+ "global_compression_ratio": null,
107
+ "ignore": [
108
+ "vision_tower.transformer.layers.0.feed_forward.gate_proj",
109
+ "vision_tower.transformer.layers.0.feed_forward.up_proj",
110
+ "vision_tower.transformer.layers.0.feed_forward.down_proj",
111
+ "vision_tower.transformer.layers.0.attention.k_proj",
112
+ "vision_tower.transformer.layers.0.attention.v_proj",
113
+ "vision_tower.transformer.layers.0.attention.q_proj",
114
+ "vision_tower.transformer.layers.0.attention.o_proj",
115
+ "vision_tower.transformer.layers.1.feed_forward.gate_proj",
116
+ "vision_tower.transformer.layers.1.feed_forward.up_proj",
117
+ "vision_tower.transformer.layers.1.feed_forward.down_proj",
118
+ "vision_tower.transformer.layers.1.attention.k_proj",
119
+ "vision_tower.transformer.layers.1.attention.v_proj",
120
+ "vision_tower.transformer.layers.1.attention.q_proj",
121
+ "vision_tower.transformer.layers.1.attention.o_proj",
122
+ "vision_tower.transformer.layers.2.feed_forward.gate_proj",
123
+ "vision_tower.transformer.layers.2.feed_forward.up_proj",
124
+ "vision_tower.transformer.layers.2.feed_forward.down_proj",
125
+ "vision_tower.transformer.layers.2.attention.k_proj",
126
+ "vision_tower.transformer.layers.2.attention.v_proj",
127
+ "vision_tower.transformer.layers.2.attention.q_proj",
128
+ "vision_tower.transformer.layers.2.attention.o_proj",
129
+ "vision_tower.transformer.layers.3.feed_forward.gate_proj",
130
+ "vision_tower.transformer.layers.3.feed_forward.up_proj",
131
+ "vision_tower.transformer.layers.3.feed_forward.down_proj",
132
+ "vision_tower.transformer.layers.3.attention.k_proj",
133
+ "vision_tower.transformer.layers.3.attention.v_proj",
134
+ "vision_tower.transformer.layers.3.attention.q_proj",
135
+ "vision_tower.transformer.layers.3.attention.o_proj",
136
+ "vision_tower.transformer.layers.4.feed_forward.gate_proj",
137
+ "vision_tower.transformer.layers.4.feed_forward.up_proj",
138
+ "vision_tower.transformer.layers.4.feed_forward.down_proj",
139
+ "vision_tower.transformer.layers.4.attention.k_proj",
140
+ "vision_tower.transformer.layers.4.attention.v_proj",
141
+ "vision_tower.transformer.layers.4.attention.q_proj",
142
+ "vision_tower.transformer.layers.4.attention.o_proj",
143
+ "vision_tower.transformer.layers.5.feed_forward.gate_proj",
144
+ "vision_tower.transformer.layers.5.feed_forward.up_proj",
145
+ "vision_tower.transformer.layers.5.feed_forward.down_proj",
146
+ "vision_tower.transformer.layers.5.attention.k_proj",
147
+ "vision_tower.transformer.layers.5.attention.v_proj",
148
+ "vision_tower.transformer.layers.5.attention.q_proj",
149
+ "vision_tower.transformer.layers.5.attention.o_proj",
150
+ "vision_tower.transformer.layers.6.feed_forward.gate_proj",
151
+ "vision_tower.transformer.layers.6.feed_forward.up_proj",
152
+ "vision_tower.transformer.layers.6.feed_forward.down_proj",
153
+ "vision_tower.transformer.layers.6.attention.k_proj",
154
+ "vision_tower.transformer.layers.6.attention.v_proj",
155
+ "vision_tower.transformer.layers.6.attention.q_proj",
156
+ "vision_tower.transformer.layers.6.attention.o_proj",
157
+ "vision_tower.transformer.layers.7.feed_forward.gate_proj",
158
+ "vision_tower.transformer.layers.7.feed_forward.up_proj",
159
+ "vision_tower.transformer.layers.7.feed_forward.down_proj",
160
+ "vision_tower.transformer.layers.7.attention.k_proj",
161
+ "vision_tower.transformer.layers.7.attention.v_proj",
162
+ "vision_tower.transformer.layers.7.attention.q_proj",
163
+ "vision_tower.transformer.layers.7.attention.o_proj",
164
+ "vision_tower.transformer.layers.8.feed_forward.gate_proj",
165
+ "vision_tower.transformer.layers.8.feed_forward.up_proj",
166
+ "vision_tower.transformer.layers.8.feed_forward.down_proj",
167
+ "vision_tower.transformer.layers.8.attention.k_proj",
168
+ "vision_tower.transformer.layers.8.attention.v_proj",
169
+ "vision_tower.transformer.layers.8.attention.q_proj",
170
+ "vision_tower.transformer.layers.8.attention.o_proj",
171
+ "vision_tower.transformer.layers.9.feed_forward.gate_proj",
172
+ "vision_tower.transformer.layers.9.feed_forward.up_proj",
173
+ "vision_tower.transformer.layers.9.feed_forward.down_proj",
174
+ "vision_tower.transformer.layers.9.attention.k_proj",
175
+ "vision_tower.transformer.layers.9.attention.v_proj",
176
+ "vision_tower.transformer.layers.9.attention.q_proj",
177
+ "vision_tower.transformer.layers.9.attention.o_proj",
178
+ "vision_tower.transformer.layers.10.feed_forward.gate_proj",
179
+ "vision_tower.transformer.layers.10.feed_forward.up_proj",
180
+ "vision_tower.transformer.layers.10.feed_forward.down_proj",
181
+ "vision_tower.transformer.layers.10.attention.k_proj",
182
+ "vision_tower.transformer.layers.10.attention.v_proj",
183
+ "vision_tower.transformer.layers.10.attention.q_proj",
184
+ "vision_tower.transformer.layers.10.attention.o_proj",
185
+ "vision_tower.transformer.layers.11.feed_forward.gate_proj",
186
+ "vision_tower.transformer.layers.11.feed_forward.up_proj",
187
+ "vision_tower.transformer.layers.11.feed_forward.down_proj",
188
+ "vision_tower.transformer.layers.11.attention.k_proj",
189
+ "vision_tower.transformer.layers.11.attention.v_proj",
190
+ "vision_tower.transformer.layers.11.attention.q_proj",
191
+ "vision_tower.transformer.layers.11.attention.o_proj",
192
+ "vision_tower.transformer.layers.12.feed_forward.gate_proj",
193
+ "vision_tower.transformer.layers.12.feed_forward.up_proj",
194
+ "vision_tower.transformer.layers.12.feed_forward.down_proj",
195
+ "vision_tower.transformer.layers.12.attention.k_proj",
196
+ "vision_tower.transformer.layers.12.attention.v_proj",
197
+ "vision_tower.transformer.layers.12.attention.q_proj",
198
+ "vision_tower.transformer.layers.12.attention.o_proj",
199
+ "vision_tower.transformer.layers.13.feed_forward.gate_proj",
200
+ "vision_tower.transformer.layers.13.feed_forward.up_proj",
201
+ "vision_tower.transformer.layers.13.feed_forward.down_proj",
202
+ "vision_tower.transformer.layers.13.attention.k_proj",
203
+ "vision_tower.transformer.layers.13.attention.v_proj",
204
+ "vision_tower.transformer.layers.13.attention.q_proj",
205
+ "vision_tower.transformer.layers.13.attention.o_proj",
206
+ "vision_tower.transformer.layers.14.feed_forward.gate_proj",
207
+ "vision_tower.transformer.layers.14.feed_forward.up_proj",
208
+ "vision_tower.transformer.layers.14.feed_forward.down_proj",
209
+ "vision_tower.transformer.layers.14.attention.k_proj",
210
+ "vision_tower.transformer.layers.14.attention.v_proj",
211
+ "vision_tower.transformer.layers.14.attention.q_proj",
212
+ "vision_tower.transformer.layers.14.attention.o_proj",
213
+ "vision_tower.transformer.layers.15.feed_forward.gate_proj",
214
+ "vision_tower.transformer.layers.15.feed_forward.up_proj",
215
+ "vision_tower.transformer.layers.15.feed_forward.down_proj",
216
+ "vision_tower.transformer.layers.15.attention.k_proj",
217
+ "vision_tower.transformer.layers.15.attention.v_proj",
218
+ "vision_tower.transformer.layers.15.attention.q_proj",
219
+ "vision_tower.transformer.layers.15.attention.o_proj",
220
+ "vision_tower.transformer.layers.16.feed_forward.gate_proj",
221
+ "vision_tower.transformer.layers.16.feed_forward.up_proj",
222
+ "vision_tower.transformer.layers.16.feed_forward.down_proj",
223
+ "vision_tower.transformer.layers.16.attention.k_proj",
224
+ "vision_tower.transformer.layers.16.attention.v_proj",
225
+ "vision_tower.transformer.layers.16.attention.q_proj",
226
+ "vision_tower.transformer.layers.16.attention.o_proj",
227
+ "vision_tower.transformer.layers.17.feed_forward.gate_proj",
228
+ "vision_tower.transformer.layers.17.feed_forward.up_proj",
229
+ "vision_tower.transformer.layers.17.feed_forward.down_proj",
230
+ "vision_tower.transformer.layers.17.attention.k_proj",
231
+ "vision_tower.transformer.layers.17.attention.v_proj",
232
+ "vision_tower.transformer.layers.17.attention.q_proj",
233
+ "vision_tower.transformer.layers.17.attention.o_proj",
234
+ "vision_tower.transformer.layers.18.feed_forward.gate_proj",
235
+ "vision_tower.transformer.layers.18.feed_forward.up_proj",
236
+ "vision_tower.transformer.layers.18.feed_forward.down_proj",
237
+ "vision_tower.transformer.layers.18.attention.k_proj",
238
+ "vision_tower.transformer.layers.18.attention.v_proj",
239
+ "vision_tower.transformer.layers.18.attention.q_proj",
240
+ "vision_tower.transformer.layers.18.attention.o_proj",
241
+ "vision_tower.transformer.layers.19.feed_forward.gate_proj",
242
+ "vision_tower.transformer.layers.19.feed_forward.up_proj",
243
+ "vision_tower.transformer.layers.19.feed_forward.down_proj",
244
+ "vision_tower.transformer.layers.19.attention.k_proj",
245
+ "vision_tower.transformer.layers.19.attention.v_proj",
246
+ "vision_tower.transformer.layers.19.attention.q_proj",
247
+ "vision_tower.transformer.layers.19.attention.o_proj",
248
+ "vision_tower.transformer.layers.20.feed_forward.gate_proj",
249
+ "vision_tower.transformer.layers.20.feed_forward.up_proj",
250
+ "vision_tower.transformer.layers.20.feed_forward.down_proj",
251
+ "vision_tower.transformer.layers.20.attention.k_proj",
252
+ "vision_tower.transformer.layers.20.attention.v_proj",
253
+ "vision_tower.transformer.layers.20.attention.q_proj",
254
+ "vision_tower.transformer.layers.20.attention.o_proj",
255
+ "vision_tower.transformer.layers.21.feed_forward.gate_proj",
256
+ "vision_tower.transformer.layers.21.feed_forward.up_proj",
257
+ "vision_tower.transformer.layers.21.feed_forward.down_proj",
258
+ "vision_tower.transformer.layers.21.attention.k_proj",
259
+ "vision_tower.transformer.layers.21.attention.v_proj",
260
+ "vision_tower.transformer.layers.21.attention.q_proj",
261
+ "vision_tower.transformer.layers.21.attention.o_proj",
262
+ "vision_tower.transformer.layers.22.feed_forward.gate_proj",
263
+ "vision_tower.transformer.layers.22.feed_forward.up_proj",
264
+ "vision_tower.transformer.layers.22.feed_forward.down_proj",
265
+ "vision_tower.transformer.layers.22.attention.k_proj",
266
+ "vision_tower.transformer.layers.22.attention.v_proj",
267
+ "vision_tower.transformer.layers.22.attention.q_proj",
268
+ "vision_tower.transformer.layers.22.attention.o_proj",
269
+ "vision_tower.transformer.layers.23.feed_forward.gate_proj",
270
+ "vision_tower.transformer.layers.23.feed_forward.up_proj",
271
+ "vision_tower.transformer.layers.23.feed_forward.down_proj",
272
+ "vision_tower.transformer.layers.23.attention.k_proj",
273
+ "vision_tower.transformer.layers.23.attention.v_proj",
274
+ "vision_tower.transformer.layers.23.attention.q_proj",
275
+ "vision_tower.transformer.layers.23.attention.o_proj",
276
+ "multi_modal_projector.patch_merger.merging_layer",
277
+ "multi_modal_projector.linear_1",
278
+ "multi_modal_projector.linear_2",
279
+ "lm_head"
280
+ ],
281
+ "kv_cache_scheme": null,
282
+ "quant_method": "compressed-tensors",
283
+ "quantization_status": "compressed",
284
+ "sparsity_config": {},
285
+ "transform_config": {},
286
+ "version": "0.19.0"
287
+ }
288
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 1,
3
+ "eos_token_id": 2,
4
+ "max_length": 262144,
5
+ "pad_token_id": 11,
6
+ "transformers_version": "5.0.0.dev0"
7
+ }
head/schema_head.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c7b61dd2b8bee2557b1384f08af37c0992963339c418ef467fa74fc3b3ee962
3
+ size 278062780
head/schema_head_config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "head": {
3
+ "hidden_size": 3072,
4
+ "n_taps": 2,
5
+ "d": 1024,
6
+ "heads": 16,
7
+ "ffn": 2816,
8
+ "route_blocks": 2,
9
+ "interact_blocks": 2,
10
+ "dropout": 0.1,
11
+ "n_roles": 5,
12
+ "ord_hidden": 256,
13
+ "max_nodes": 4096,
14
+ "grad_checkpoint": true,
15
+ "taps": [
16
+ "final",
17
+ 19
18
+ ],
19
+ "template": "schema-v1"
20
+ },
21
+ "taps": [
22
+ "final",
23
+ 19
24
+ ],
25
+ "template": "schema-v1-nosys",
26
+ "step": 3361,
27
+ "lora_rank": 64,
28
+ "snapshot": "StandardThinking/StandardOne-3B-SH",
29
+ "preview": true
30
+ }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32c9aa6a15284a7491546f5d69fe8a12af4bab27f7592ca596c8eec807f95de6
3
+ size 2901732848
model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d253eb9a626fde1efa54ec898bd4eb20afb1774c814269582411197e412275aa
3
+ size 1771938896
model.safetensors.index.json ADDED
@@ -0,0 +1,647 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 4673671744
4
+ },
5
+ "weight_map": {
6
+ "language_model.model.embed_tokens.weight": "model-00001-of-00002.safetensors",
7
+ "language_model.model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
8
+ "language_model.model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
9
+ "language_model.model.layers.0.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
10
+ "language_model.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
11
+ "language_model.model.layers.0.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
12
+ "language_model.model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
13
+ "language_model.model.layers.0.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
14
+ "language_model.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
15
+ "language_model.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
16
+ "language_model.model.layers.0.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
17
+ "language_model.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
18
+ "language_model.model.layers.0.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
19
+ "language_model.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
20
+ "language_model.model.layers.0.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
21
+ "language_model.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
22
+ "language_model.model.layers.0.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
23
+ "language_model.model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
24
+ "language_model.model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
25
+ "language_model.model.layers.1.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
26
+ "language_model.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
27
+ "language_model.model.layers.1.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
28
+ "language_model.model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
29
+ "language_model.model.layers.1.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
30
+ "language_model.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
31
+ "language_model.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
32
+ "language_model.model.layers.1.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
33
+ "language_model.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
34
+ "language_model.model.layers.1.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
35
+ "language_model.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
36
+ "language_model.model.layers.1.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
37
+ "language_model.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
38
+ "language_model.model.layers.1.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
39
+ "language_model.model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
40
+ "language_model.model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
41
+ "language_model.model.layers.10.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
42
+ "language_model.model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
43
+ "language_model.model.layers.10.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
44
+ "language_model.model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
45
+ "language_model.model.layers.10.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
46
+ "language_model.model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
47
+ "language_model.model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
48
+ "language_model.model.layers.10.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
49
+ "language_model.model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
50
+ "language_model.model.layers.10.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
51
+ "language_model.model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
52
+ "language_model.model.layers.10.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
53
+ "language_model.model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
54
+ "language_model.model.layers.10.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
55
+ "language_model.model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
56
+ "language_model.model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
57
+ "language_model.model.layers.11.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
58
+ "language_model.model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
59
+ "language_model.model.layers.11.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
60
+ "language_model.model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
61
+ "language_model.model.layers.11.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
62
+ "language_model.model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
63
+ "language_model.model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
64
+ "language_model.model.layers.11.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
65
+ "language_model.model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
66
+ "language_model.model.layers.11.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
67
+ "language_model.model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
68
+ "language_model.model.layers.11.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
69
+ "language_model.model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
70
+ "language_model.model.layers.11.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
71
+ "language_model.model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
72
+ "language_model.model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
73
+ "language_model.model.layers.12.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
74
+ "language_model.model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
75
+ "language_model.model.layers.12.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
76
+ "language_model.model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
77
+ "language_model.model.layers.12.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
78
+ "language_model.model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
79
+ "language_model.model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
80
+ "language_model.model.layers.12.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
81
+ "language_model.model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
82
+ "language_model.model.layers.12.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
83
+ "language_model.model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
84
+ "language_model.model.layers.12.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
85
+ "language_model.model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
86
+ "language_model.model.layers.12.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
87
+ "language_model.model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
88
+ "language_model.model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
89
+ "language_model.model.layers.13.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
90
+ "language_model.model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
91
+ "language_model.model.layers.13.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
92
+ "language_model.model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
93
+ "language_model.model.layers.13.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
94
+ "language_model.model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
95
+ "language_model.model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
96
+ "language_model.model.layers.13.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
97
+ "language_model.model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
98
+ "language_model.model.layers.13.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
99
+ "language_model.model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
100
+ "language_model.model.layers.13.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
101
+ "language_model.model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
102
+ "language_model.model.layers.13.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
103
+ "language_model.model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
104
+ "language_model.model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
105
+ "language_model.model.layers.14.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
106
+ "language_model.model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
107
+ "language_model.model.layers.14.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
108
+ "language_model.model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
109
+ "language_model.model.layers.14.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
110
+ "language_model.model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
111
+ "language_model.model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
112
+ "language_model.model.layers.14.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
113
+ "language_model.model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
114
+ "language_model.model.layers.14.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
115
+ "language_model.model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
116
+ "language_model.model.layers.14.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
117
+ "language_model.model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
118
+ "language_model.model.layers.14.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
119
+ "language_model.model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
120
+ "language_model.model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
121
+ "language_model.model.layers.15.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
122
+ "language_model.model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
123
+ "language_model.model.layers.15.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
124
+ "language_model.model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
125
+ "language_model.model.layers.15.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
126
+ "language_model.model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
127
+ "language_model.model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
128
+ "language_model.model.layers.15.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
129
+ "language_model.model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
130
+ "language_model.model.layers.15.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
131
+ "language_model.model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
132
+ "language_model.model.layers.15.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
133
+ "language_model.model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
134
+ "language_model.model.layers.15.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
135
+ "language_model.model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
136
+ "language_model.model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
137
+ "language_model.model.layers.16.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
138
+ "language_model.model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
139
+ "language_model.model.layers.16.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
140
+ "language_model.model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
141
+ "language_model.model.layers.16.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
142
+ "language_model.model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
143
+ "language_model.model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
144
+ "language_model.model.layers.16.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
145
+ "language_model.model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
146
+ "language_model.model.layers.16.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
147
+ "language_model.model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
148
+ "language_model.model.layers.16.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
149
+ "language_model.model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
150
+ "language_model.model.layers.16.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
151
+ "language_model.model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
152
+ "language_model.model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
153
+ "language_model.model.layers.17.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
154
+ "language_model.model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
155
+ "language_model.model.layers.17.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
156
+ "language_model.model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
157
+ "language_model.model.layers.17.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
158
+ "language_model.model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
159
+ "language_model.model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
160
+ "language_model.model.layers.17.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
161
+ "language_model.model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
162
+ "language_model.model.layers.17.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
163
+ "language_model.model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
164
+ "language_model.model.layers.17.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
165
+ "language_model.model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
166
+ "language_model.model.layers.17.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
167
+ "language_model.model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
168
+ "language_model.model.layers.18.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
169
+ "language_model.model.layers.18.mlp.down_proj.weight_scale": "model-00002-of-00002.safetensors",
170
+ "language_model.model.layers.18.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
171
+ "language_model.model.layers.18.mlp.gate_proj.weight_scale": "model-00002-of-00002.safetensors",
172
+ "language_model.model.layers.18.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
173
+ "language_model.model.layers.18.mlp.up_proj.weight_scale": "model-00002-of-00002.safetensors",
174
+ "language_model.model.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
175
+ "language_model.model.layers.18.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
176
+ "language_model.model.layers.18.self_attn.k_proj.weight_scale": "model-00002-of-00002.safetensors",
177
+ "language_model.model.layers.18.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
178
+ "language_model.model.layers.18.self_attn.o_proj.weight_scale": "model-00002-of-00002.safetensors",
179
+ "language_model.model.layers.18.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
180
+ "language_model.model.layers.18.self_attn.q_proj.weight_scale": "model-00002-of-00002.safetensors",
181
+ "language_model.model.layers.18.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
182
+ "language_model.model.layers.18.self_attn.v_proj.weight_scale": "model-00002-of-00002.safetensors",
183
+ "language_model.model.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors",
184
+ "language_model.model.layers.19.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
185
+ "language_model.model.layers.19.mlp.down_proj.weight_scale": "model-00002-of-00002.safetensors",
186
+ "language_model.model.layers.19.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
187
+ "language_model.model.layers.19.mlp.gate_proj.weight_scale": "model-00002-of-00002.safetensors",
188
+ "language_model.model.layers.19.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
189
+ "language_model.model.layers.19.mlp.up_proj.weight_scale": "model-00002-of-00002.safetensors",
190
+ "language_model.model.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
191
+ "language_model.model.layers.19.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
192
+ "language_model.model.layers.19.self_attn.k_proj.weight_scale": "model-00002-of-00002.safetensors",
193
+ "language_model.model.layers.19.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
194
+ "language_model.model.layers.19.self_attn.o_proj.weight_scale": "model-00002-of-00002.safetensors",
195
+ "language_model.model.layers.19.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
196
+ "language_model.model.layers.19.self_attn.q_proj.weight_scale": "model-00002-of-00002.safetensors",
197
+ "language_model.model.layers.19.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
198
+ "language_model.model.layers.19.self_attn.v_proj.weight_scale": "model-00002-of-00002.safetensors",
199
+ "language_model.model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
200
+ "language_model.model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
201
+ "language_model.model.layers.2.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
202
+ "language_model.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
203
+ "language_model.model.layers.2.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
204
+ "language_model.model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
205
+ "language_model.model.layers.2.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
206
+ "language_model.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
207
+ "language_model.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
208
+ "language_model.model.layers.2.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
209
+ "language_model.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
210
+ "language_model.model.layers.2.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
211
+ "language_model.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
212
+ "language_model.model.layers.2.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
213
+ "language_model.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
214
+ "language_model.model.layers.2.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
215
+ "language_model.model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors",
216
+ "language_model.model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
217
+ "language_model.model.layers.20.mlp.down_proj.weight_scale": "model-00002-of-00002.safetensors",
218
+ "language_model.model.layers.20.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
219
+ "language_model.model.layers.20.mlp.gate_proj.weight_scale": "model-00002-of-00002.safetensors",
220
+ "language_model.model.layers.20.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
221
+ "language_model.model.layers.20.mlp.up_proj.weight_scale": "model-00002-of-00002.safetensors",
222
+ "language_model.model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
223
+ "language_model.model.layers.20.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
224
+ "language_model.model.layers.20.self_attn.k_proj.weight_scale": "model-00002-of-00002.safetensors",
225
+ "language_model.model.layers.20.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
226
+ "language_model.model.layers.20.self_attn.o_proj.weight_scale": "model-00002-of-00002.safetensors",
227
+ "language_model.model.layers.20.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
228
+ "language_model.model.layers.20.self_attn.q_proj.weight_scale": "model-00002-of-00002.safetensors",
229
+ "language_model.model.layers.20.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
230
+ "language_model.model.layers.20.self_attn.v_proj.weight_scale": "model-00002-of-00002.safetensors",
231
+ "language_model.model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
232
+ "language_model.model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
233
+ "language_model.model.layers.21.mlp.down_proj.weight_scale": "model-00002-of-00002.safetensors",
234
+ "language_model.model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
235
+ "language_model.model.layers.21.mlp.gate_proj.weight_scale": "model-00002-of-00002.safetensors",
236
+ "language_model.model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
237
+ "language_model.model.layers.21.mlp.up_proj.weight_scale": "model-00002-of-00002.safetensors",
238
+ "language_model.model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
239
+ "language_model.model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
240
+ "language_model.model.layers.21.self_attn.k_proj.weight_scale": "model-00002-of-00002.safetensors",
241
+ "language_model.model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
242
+ "language_model.model.layers.21.self_attn.o_proj.weight_scale": "model-00002-of-00002.safetensors",
243
+ "language_model.model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
244
+ "language_model.model.layers.21.self_attn.q_proj.weight_scale": "model-00002-of-00002.safetensors",
245
+ "language_model.model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
246
+ "language_model.model.layers.21.self_attn.v_proj.weight_scale": "model-00002-of-00002.safetensors",
247
+ "language_model.model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
248
+ "language_model.model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
249
+ "language_model.model.layers.22.mlp.down_proj.weight_scale": "model-00002-of-00002.safetensors",
250
+ "language_model.model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
251
+ "language_model.model.layers.22.mlp.gate_proj.weight_scale": "model-00002-of-00002.safetensors",
252
+ "language_model.model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
253
+ "language_model.model.layers.22.mlp.up_proj.weight_scale": "model-00002-of-00002.safetensors",
254
+ "language_model.model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
255
+ "language_model.model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
256
+ "language_model.model.layers.22.self_attn.k_proj.weight_scale": "model-00002-of-00002.safetensors",
257
+ "language_model.model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
258
+ "language_model.model.layers.22.self_attn.o_proj.weight_scale": "model-00002-of-00002.safetensors",
259
+ "language_model.model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
260
+ "language_model.model.layers.22.self_attn.q_proj.weight_scale": "model-00002-of-00002.safetensors",
261
+ "language_model.model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
262
+ "language_model.model.layers.22.self_attn.v_proj.weight_scale": "model-00002-of-00002.safetensors",
263
+ "language_model.model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
264
+ "language_model.model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
265
+ "language_model.model.layers.23.mlp.down_proj.weight_scale": "model-00002-of-00002.safetensors",
266
+ "language_model.model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
267
+ "language_model.model.layers.23.mlp.gate_proj.weight_scale": "model-00002-of-00002.safetensors",
268
+ "language_model.model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
269
+ "language_model.model.layers.23.mlp.up_proj.weight_scale": "model-00002-of-00002.safetensors",
270
+ "language_model.model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
271
+ "language_model.model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
272
+ "language_model.model.layers.23.self_attn.k_proj.weight_scale": "model-00002-of-00002.safetensors",
273
+ "language_model.model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
274
+ "language_model.model.layers.23.self_attn.o_proj.weight_scale": "model-00002-of-00002.safetensors",
275
+ "language_model.model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
276
+ "language_model.model.layers.23.self_attn.q_proj.weight_scale": "model-00002-of-00002.safetensors",
277
+ "language_model.model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
278
+ "language_model.model.layers.23.self_attn.v_proj.weight_scale": "model-00002-of-00002.safetensors",
279
+ "language_model.model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
280
+ "language_model.model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
281
+ "language_model.model.layers.24.mlp.down_proj.weight_scale": "model-00002-of-00002.safetensors",
282
+ "language_model.model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
283
+ "language_model.model.layers.24.mlp.gate_proj.weight_scale": "model-00002-of-00002.safetensors",
284
+ "language_model.model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
285
+ "language_model.model.layers.24.mlp.up_proj.weight_scale": "model-00002-of-00002.safetensors",
286
+ "language_model.model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
287
+ "language_model.model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
288
+ "language_model.model.layers.24.self_attn.k_proj.weight_scale": "model-00002-of-00002.safetensors",
289
+ "language_model.model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
290
+ "language_model.model.layers.24.self_attn.o_proj.weight_scale": "model-00002-of-00002.safetensors",
291
+ "language_model.model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
292
+ "language_model.model.layers.24.self_attn.q_proj.weight_scale": "model-00002-of-00002.safetensors",
293
+ "language_model.model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
294
+ "language_model.model.layers.24.self_attn.v_proj.weight_scale": "model-00002-of-00002.safetensors",
295
+ "language_model.model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
296
+ "language_model.model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
297
+ "language_model.model.layers.25.mlp.down_proj.weight_scale": "model-00002-of-00002.safetensors",
298
+ "language_model.model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
299
+ "language_model.model.layers.25.mlp.gate_proj.weight_scale": "model-00002-of-00002.safetensors",
300
+ "language_model.model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
301
+ "language_model.model.layers.25.mlp.up_proj.weight_scale": "model-00002-of-00002.safetensors",
302
+ "language_model.model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
303
+ "language_model.model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
304
+ "language_model.model.layers.25.self_attn.k_proj.weight_scale": "model-00002-of-00002.safetensors",
305
+ "language_model.model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
306
+ "language_model.model.layers.25.self_attn.o_proj.weight_scale": "model-00002-of-00002.safetensors",
307
+ "language_model.model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
308
+ "language_model.model.layers.25.self_attn.q_proj.weight_scale": "model-00002-of-00002.safetensors",
309
+ "language_model.model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
310
+ "language_model.model.layers.25.self_attn.v_proj.weight_scale": "model-00002-of-00002.safetensors",
311
+ "language_model.model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
312
+ "language_model.model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
313
+ "language_model.model.layers.3.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
314
+ "language_model.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
315
+ "language_model.model.layers.3.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
316
+ "language_model.model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
317
+ "language_model.model.layers.3.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
318
+ "language_model.model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
319
+ "language_model.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
320
+ "language_model.model.layers.3.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
321
+ "language_model.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
322
+ "language_model.model.layers.3.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
323
+ "language_model.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
324
+ "language_model.model.layers.3.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
325
+ "language_model.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
326
+ "language_model.model.layers.3.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
327
+ "language_model.model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
328
+ "language_model.model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
329
+ "language_model.model.layers.4.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
330
+ "language_model.model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
331
+ "language_model.model.layers.4.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
332
+ "language_model.model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
333
+ "language_model.model.layers.4.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
334
+ "language_model.model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
335
+ "language_model.model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
336
+ "language_model.model.layers.4.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
337
+ "language_model.model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
338
+ "language_model.model.layers.4.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
339
+ "language_model.model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
340
+ "language_model.model.layers.4.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
341
+ "language_model.model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
342
+ "language_model.model.layers.4.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
343
+ "language_model.model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
344
+ "language_model.model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
345
+ "language_model.model.layers.5.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
346
+ "language_model.model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
347
+ "language_model.model.layers.5.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
348
+ "language_model.model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
349
+ "language_model.model.layers.5.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
350
+ "language_model.model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
351
+ "language_model.model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
352
+ "language_model.model.layers.5.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
353
+ "language_model.model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
354
+ "language_model.model.layers.5.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
355
+ "language_model.model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
356
+ "language_model.model.layers.5.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
357
+ "language_model.model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
358
+ "language_model.model.layers.5.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
359
+ "language_model.model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
360
+ "language_model.model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
361
+ "language_model.model.layers.6.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
362
+ "language_model.model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
363
+ "language_model.model.layers.6.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
364
+ "language_model.model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
365
+ "language_model.model.layers.6.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
366
+ "language_model.model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
367
+ "language_model.model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
368
+ "language_model.model.layers.6.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
369
+ "language_model.model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
370
+ "language_model.model.layers.6.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
371
+ "language_model.model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
372
+ "language_model.model.layers.6.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
373
+ "language_model.model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
374
+ "language_model.model.layers.6.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
375
+ "language_model.model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
376
+ "language_model.model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
377
+ "language_model.model.layers.7.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
378
+ "language_model.model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
379
+ "language_model.model.layers.7.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
380
+ "language_model.model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
381
+ "language_model.model.layers.7.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
382
+ "language_model.model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
383
+ "language_model.model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
384
+ "language_model.model.layers.7.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
385
+ "language_model.model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
386
+ "language_model.model.layers.7.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
387
+ "language_model.model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
388
+ "language_model.model.layers.7.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
389
+ "language_model.model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
390
+ "language_model.model.layers.7.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
391
+ "language_model.model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
392
+ "language_model.model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
393
+ "language_model.model.layers.8.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
394
+ "language_model.model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
395
+ "language_model.model.layers.8.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
396
+ "language_model.model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
397
+ "language_model.model.layers.8.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
398
+ "language_model.model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
399
+ "language_model.model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
400
+ "language_model.model.layers.8.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
401
+ "language_model.model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
402
+ "language_model.model.layers.8.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
403
+ "language_model.model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
404
+ "language_model.model.layers.8.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
405
+ "language_model.model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
406
+ "language_model.model.layers.8.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
407
+ "language_model.model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
408
+ "language_model.model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
409
+ "language_model.model.layers.9.mlp.down_proj.weight_scale": "model-00001-of-00002.safetensors",
410
+ "language_model.model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
411
+ "language_model.model.layers.9.mlp.gate_proj.weight_scale": "model-00001-of-00002.safetensors",
412
+ "language_model.model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
413
+ "language_model.model.layers.9.mlp.up_proj.weight_scale": "model-00001-of-00002.safetensors",
414
+ "language_model.model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
415
+ "language_model.model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
416
+ "language_model.model.layers.9.self_attn.k_proj.weight_scale": "model-00001-of-00002.safetensors",
417
+ "language_model.model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
418
+ "language_model.model.layers.9.self_attn.o_proj.weight_scale": "model-00001-of-00002.safetensors",
419
+ "language_model.model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
420
+ "language_model.model.layers.9.self_attn.q_proj.weight_scale": "model-00001-of-00002.safetensors",
421
+ "language_model.model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
422
+ "language_model.model.layers.9.self_attn.v_proj.weight_scale": "model-00001-of-00002.safetensors",
423
+ "language_model.model.norm.weight": "model-00002-of-00002.safetensors",
424
+ "multi_modal_projector.linear_1.weight": "model-00002-of-00002.safetensors",
425
+ "multi_modal_projector.linear_2.weight": "model-00002-of-00002.safetensors",
426
+ "multi_modal_projector.norm.weight": "model-00002-of-00002.safetensors",
427
+ "multi_modal_projector.patch_merger.merging_layer.weight": "model-00002-of-00002.safetensors",
428
+ "vision_tower.ln_pre.weight": "model-00002-of-00002.safetensors",
429
+ "vision_tower.patch_conv.weight": "model-00002-of-00002.safetensors",
430
+ "vision_tower.transformer.layers.0.attention.k_proj.weight": "model-00002-of-00002.safetensors",
431
+ "vision_tower.transformer.layers.0.attention.o_proj.weight": "model-00002-of-00002.safetensors",
432
+ "vision_tower.transformer.layers.0.attention.q_proj.weight": "model-00002-of-00002.safetensors",
433
+ "vision_tower.transformer.layers.0.attention.v_proj.weight": "model-00002-of-00002.safetensors",
434
+ "vision_tower.transformer.layers.0.attention_norm.weight": "model-00002-of-00002.safetensors",
435
+ "vision_tower.transformer.layers.0.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
436
+ "vision_tower.transformer.layers.0.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
437
+ "vision_tower.transformer.layers.0.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
438
+ "vision_tower.transformer.layers.0.ffn_norm.weight": "model-00002-of-00002.safetensors",
439
+ "vision_tower.transformer.layers.1.attention.k_proj.weight": "model-00002-of-00002.safetensors",
440
+ "vision_tower.transformer.layers.1.attention.o_proj.weight": "model-00002-of-00002.safetensors",
441
+ "vision_tower.transformer.layers.1.attention.q_proj.weight": "model-00002-of-00002.safetensors",
442
+ "vision_tower.transformer.layers.1.attention.v_proj.weight": "model-00002-of-00002.safetensors",
443
+ "vision_tower.transformer.layers.1.attention_norm.weight": "model-00002-of-00002.safetensors",
444
+ "vision_tower.transformer.layers.1.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
445
+ "vision_tower.transformer.layers.1.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
446
+ "vision_tower.transformer.layers.1.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
447
+ "vision_tower.transformer.layers.1.ffn_norm.weight": "model-00002-of-00002.safetensors",
448
+ "vision_tower.transformer.layers.10.attention.k_proj.weight": "model-00002-of-00002.safetensors",
449
+ "vision_tower.transformer.layers.10.attention.o_proj.weight": "model-00002-of-00002.safetensors",
450
+ "vision_tower.transformer.layers.10.attention.q_proj.weight": "model-00002-of-00002.safetensors",
451
+ "vision_tower.transformer.layers.10.attention.v_proj.weight": "model-00002-of-00002.safetensors",
452
+ "vision_tower.transformer.layers.10.attention_norm.weight": "model-00002-of-00002.safetensors",
453
+ "vision_tower.transformer.layers.10.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
454
+ "vision_tower.transformer.layers.10.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
455
+ "vision_tower.transformer.layers.10.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
456
+ "vision_tower.transformer.layers.10.ffn_norm.weight": "model-00002-of-00002.safetensors",
457
+ "vision_tower.transformer.layers.11.attention.k_proj.weight": "model-00002-of-00002.safetensors",
458
+ "vision_tower.transformer.layers.11.attention.o_proj.weight": "model-00002-of-00002.safetensors",
459
+ "vision_tower.transformer.layers.11.attention.q_proj.weight": "model-00002-of-00002.safetensors",
460
+ "vision_tower.transformer.layers.11.attention.v_proj.weight": "model-00002-of-00002.safetensors",
461
+ "vision_tower.transformer.layers.11.attention_norm.weight": "model-00002-of-00002.safetensors",
462
+ "vision_tower.transformer.layers.11.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
463
+ "vision_tower.transformer.layers.11.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
464
+ "vision_tower.transformer.layers.11.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
465
+ "vision_tower.transformer.layers.11.ffn_norm.weight": "model-00002-of-00002.safetensors",
466
+ "vision_tower.transformer.layers.12.attention.k_proj.weight": "model-00002-of-00002.safetensors",
467
+ "vision_tower.transformer.layers.12.attention.o_proj.weight": "model-00002-of-00002.safetensors",
468
+ "vision_tower.transformer.layers.12.attention.q_proj.weight": "model-00002-of-00002.safetensors",
469
+ "vision_tower.transformer.layers.12.attention.v_proj.weight": "model-00002-of-00002.safetensors",
470
+ "vision_tower.transformer.layers.12.attention_norm.weight": "model-00002-of-00002.safetensors",
471
+ "vision_tower.transformer.layers.12.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
472
+ "vision_tower.transformer.layers.12.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
473
+ "vision_tower.transformer.layers.12.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
474
+ "vision_tower.transformer.layers.12.ffn_norm.weight": "model-00002-of-00002.safetensors",
475
+ "vision_tower.transformer.layers.13.attention.k_proj.weight": "model-00002-of-00002.safetensors",
476
+ "vision_tower.transformer.layers.13.attention.o_proj.weight": "model-00002-of-00002.safetensors",
477
+ "vision_tower.transformer.layers.13.attention.q_proj.weight": "model-00002-of-00002.safetensors",
478
+ "vision_tower.transformer.layers.13.attention.v_proj.weight": "model-00002-of-00002.safetensors",
479
+ "vision_tower.transformer.layers.13.attention_norm.weight": "model-00002-of-00002.safetensors",
480
+ "vision_tower.transformer.layers.13.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
481
+ "vision_tower.transformer.layers.13.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
482
+ "vision_tower.transformer.layers.13.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
483
+ "vision_tower.transformer.layers.13.ffn_norm.weight": "model-00002-of-00002.safetensors",
484
+ "vision_tower.transformer.layers.14.attention.k_proj.weight": "model-00002-of-00002.safetensors",
485
+ "vision_tower.transformer.layers.14.attention.o_proj.weight": "model-00002-of-00002.safetensors",
486
+ "vision_tower.transformer.layers.14.attention.q_proj.weight": "model-00002-of-00002.safetensors",
487
+ "vision_tower.transformer.layers.14.attention.v_proj.weight": "model-00002-of-00002.safetensors",
488
+ "vision_tower.transformer.layers.14.attention_norm.weight": "model-00002-of-00002.safetensors",
489
+ "vision_tower.transformer.layers.14.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
490
+ "vision_tower.transformer.layers.14.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
491
+ "vision_tower.transformer.layers.14.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
492
+ "vision_tower.transformer.layers.14.ffn_norm.weight": "model-00002-of-00002.safetensors",
493
+ "vision_tower.transformer.layers.15.attention.k_proj.weight": "model-00002-of-00002.safetensors",
494
+ "vision_tower.transformer.layers.15.attention.o_proj.weight": "model-00002-of-00002.safetensors",
495
+ "vision_tower.transformer.layers.15.attention.q_proj.weight": "model-00002-of-00002.safetensors",
496
+ "vision_tower.transformer.layers.15.attention.v_proj.weight": "model-00002-of-00002.safetensors",
497
+ "vision_tower.transformer.layers.15.attention_norm.weight": "model-00002-of-00002.safetensors",
498
+ "vision_tower.transformer.layers.15.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
499
+ "vision_tower.transformer.layers.15.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
500
+ "vision_tower.transformer.layers.15.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
501
+ "vision_tower.transformer.layers.15.ffn_norm.weight": "model-00002-of-00002.safetensors",
502
+ "vision_tower.transformer.layers.16.attention.k_proj.weight": "model-00002-of-00002.safetensors",
503
+ "vision_tower.transformer.layers.16.attention.o_proj.weight": "model-00002-of-00002.safetensors",
504
+ "vision_tower.transformer.layers.16.attention.q_proj.weight": "model-00002-of-00002.safetensors",
505
+ "vision_tower.transformer.layers.16.attention.v_proj.weight": "model-00002-of-00002.safetensors",
506
+ "vision_tower.transformer.layers.16.attention_norm.weight": "model-00002-of-00002.safetensors",
507
+ "vision_tower.transformer.layers.16.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
508
+ "vision_tower.transformer.layers.16.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
509
+ "vision_tower.transformer.layers.16.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
510
+ "vision_tower.transformer.layers.16.ffn_norm.weight": "model-00002-of-00002.safetensors",
511
+ "vision_tower.transformer.layers.17.attention.k_proj.weight": "model-00002-of-00002.safetensors",
512
+ "vision_tower.transformer.layers.17.attention.o_proj.weight": "model-00002-of-00002.safetensors",
513
+ "vision_tower.transformer.layers.17.attention.q_proj.weight": "model-00002-of-00002.safetensors",
514
+ "vision_tower.transformer.layers.17.attention.v_proj.weight": "model-00002-of-00002.safetensors",
515
+ "vision_tower.transformer.layers.17.attention_norm.weight": "model-00002-of-00002.safetensors",
516
+ "vision_tower.transformer.layers.17.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
517
+ "vision_tower.transformer.layers.17.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
518
+ "vision_tower.transformer.layers.17.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
519
+ "vision_tower.transformer.layers.17.ffn_norm.weight": "model-00002-of-00002.safetensors",
520
+ "vision_tower.transformer.layers.18.attention.k_proj.weight": "model-00002-of-00002.safetensors",
521
+ "vision_tower.transformer.layers.18.attention.o_proj.weight": "model-00002-of-00002.safetensors",
522
+ "vision_tower.transformer.layers.18.attention.q_proj.weight": "model-00002-of-00002.safetensors",
523
+ "vision_tower.transformer.layers.18.attention.v_proj.weight": "model-00002-of-00002.safetensors",
524
+ "vision_tower.transformer.layers.18.attention_norm.weight": "model-00002-of-00002.safetensors",
525
+ "vision_tower.transformer.layers.18.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
526
+ "vision_tower.transformer.layers.18.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
527
+ "vision_tower.transformer.layers.18.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
528
+ "vision_tower.transformer.layers.18.ffn_norm.weight": "model-00002-of-00002.safetensors",
529
+ "vision_tower.transformer.layers.19.attention.k_proj.weight": "model-00002-of-00002.safetensors",
530
+ "vision_tower.transformer.layers.19.attention.o_proj.weight": "model-00002-of-00002.safetensors",
531
+ "vision_tower.transformer.layers.19.attention.q_proj.weight": "model-00002-of-00002.safetensors",
532
+ "vision_tower.transformer.layers.19.attention.v_proj.weight": "model-00002-of-00002.safetensors",
533
+ "vision_tower.transformer.layers.19.attention_norm.weight": "model-00002-of-00002.safetensors",
534
+ "vision_tower.transformer.layers.19.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
535
+ "vision_tower.transformer.layers.19.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
536
+ "vision_tower.transformer.layers.19.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
537
+ "vision_tower.transformer.layers.19.ffn_norm.weight": "model-00002-of-00002.safetensors",
538
+ "vision_tower.transformer.layers.2.attention.k_proj.weight": "model-00002-of-00002.safetensors",
539
+ "vision_tower.transformer.layers.2.attention.o_proj.weight": "model-00002-of-00002.safetensors",
540
+ "vision_tower.transformer.layers.2.attention.q_proj.weight": "model-00002-of-00002.safetensors",
541
+ "vision_tower.transformer.layers.2.attention.v_proj.weight": "model-00002-of-00002.safetensors",
542
+ "vision_tower.transformer.layers.2.attention_norm.weight": "model-00002-of-00002.safetensors",
543
+ "vision_tower.transformer.layers.2.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
544
+ "vision_tower.transformer.layers.2.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
545
+ "vision_tower.transformer.layers.2.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
546
+ "vision_tower.transformer.layers.2.ffn_norm.weight": "model-00002-of-00002.safetensors",
547
+ "vision_tower.transformer.layers.20.attention.k_proj.weight": "model-00002-of-00002.safetensors",
548
+ "vision_tower.transformer.layers.20.attention.o_proj.weight": "model-00002-of-00002.safetensors",
549
+ "vision_tower.transformer.layers.20.attention.q_proj.weight": "model-00002-of-00002.safetensors",
550
+ "vision_tower.transformer.layers.20.attention.v_proj.weight": "model-00002-of-00002.safetensors",
551
+ "vision_tower.transformer.layers.20.attention_norm.weight": "model-00002-of-00002.safetensors",
552
+ "vision_tower.transformer.layers.20.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
553
+ "vision_tower.transformer.layers.20.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
554
+ "vision_tower.transformer.layers.20.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
555
+ "vision_tower.transformer.layers.20.ffn_norm.weight": "model-00002-of-00002.safetensors",
556
+ "vision_tower.transformer.layers.21.attention.k_proj.weight": "model-00002-of-00002.safetensors",
557
+ "vision_tower.transformer.layers.21.attention.o_proj.weight": "model-00002-of-00002.safetensors",
558
+ "vision_tower.transformer.layers.21.attention.q_proj.weight": "model-00002-of-00002.safetensors",
559
+ "vision_tower.transformer.layers.21.attention.v_proj.weight": "model-00002-of-00002.safetensors",
560
+ "vision_tower.transformer.layers.21.attention_norm.weight": "model-00002-of-00002.safetensors",
561
+ "vision_tower.transformer.layers.21.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
562
+ "vision_tower.transformer.layers.21.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
563
+ "vision_tower.transformer.layers.21.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
564
+ "vision_tower.transformer.layers.21.ffn_norm.weight": "model-00002-of-00002.safetensors",
565
+ "vision_tower.transformer.layers.22.attention.k_proj.weight": "model-00002-of-00002.safetensors",
566
+ "vision_tower.transformer.layers.22.attention.o_proj.weight": "model-00002-of-00002.safetensors",
567
+ "vision_tower.transformer.layers.22.attention.q_proj.weight": "model-00002-of-00002.safetensors",
568
+ "vision_tower.transformer.layers.22.attention.v_proj.weight": "model-00002-of-00002.safetensors",
569
+ "vision_tower.transformer.layers.22.attention_norm.weight": "model-00002-of-00002.safetensors",
570
+ "vision_tower.transformer.layers.22.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
571
+ "vision_tower.transformer.layers.22.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
572
+ "vision_tower.transformer.layers.22.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
573
+ "vision_tower.transformer.layers.22.ffn_norm.weight": "model-00002-of-00002.safetensors",
574
+ "vision_tower.transformer.layers.23.attention.k_proj.weight": "model-00002-of-00002.safetensors",
575
+ "vision_tower.transformer.layers.23.attention.o_proj.weight": "model-00002-of-00002.safetensors",
576
+ "vision_tower.transformer.layers.23.attention.q_proj.weight": "model-00002-of-00002.safetensors",
577
+ "vision_tower.transformer.layers.23.attention.v_proj.weight": "model-00002-of-00002.safetensors",
578
+ "vision_tower.transformer.layers.23.attention_norm.weight": "model-00002-of-00002.safetensors",
579
+ "vision_tower.transformer.layers.23.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
580
+ "vision_tower.transformer.layers.23.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
581
+ "vision_tower.transformer.layers.23.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
582
+ "vision_tower.transformer.layers.23.ffn_norm.weight": "model-00002-of-00002.safetensors",
583
+ "vision_tower.transformer.layers.3.attention.k_proj.weight": "model-00002-of-00002.safetensors",
584
+ "vision_tower.transformer.layers.3.attention.o_proj.weight": "model-00002-of-00002.safetensors",
585
+ "vision_tower.transformer.layers.3.attention.q_proj.weight": "model-00002-of-00002.safetensors",
586
+ "vision_tower.transformer.layers.3.attention.v_proj.weight": "model-00002-of-00002.safetensors",
587
+ "vision_tower.transformer.layers.3.attention_norm.weight": "model-00002-of-00002.safetensors",
588
+ "vision_tower.transformer.layers.3.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
589
+ "vision_tower.transformer.layers.3.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
590
+ "vision_tower.transformer.layers.3.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
591
+ "vision_tower.transformer.layers.3.ffn_norm.weight": "model-00002-of-00002.safetensors",
592
+ "vision_tower.transformer.layers.4.attention.k_proj.weight": "model-00002-of-00002.safetensors",
593
+ "vision_tower.transformer.layers.4.attention.o_proj.weight": "model-00002-of-00002.safetensors",
594
+ "vision_tower.transformer.layers.4.attention.q_proj.weight": "model-00002-of-00002.safetensors",
595
+ "vision_tower.transformer.layers.4.attention.v_proj.weight": "model-00002-of-00002.safetensors",
596
+ "vision_tower.transformer.layers.4.attention_norm.weight": "model-00002-of-00002.safetensors",
597
+ "vision_tower.transformer.layers.4.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
598
+ "vision_tower.transformer.layers.4.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
599
+ "vision_tower.transformer.layers.4.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
600
+ "vision_tower.transformer.layers.4.ffn_norm.weight": "model-00002-of-00002.safetensors",
601
+ "vision_tower.transformer.layers.5.attention.k_proj.weight": "model-00002-of-00002.safetensors",
602
+ "vision_tower.transformer.layers.5.attention.o_proj.weight": "model-00002-of-00002.safetensors",
603
+ "vision_tower.transformer.layers.5.attention.q_proj.weight": "model-00002-of-00002.safetensors",
604
+ "vision_tower.transformer.layers.5.attention.v_proj.weight": "model-00002-of-00002.safetensors",
605
+ "vision_tower.transformer.layers.5.attention_norm.weight": "model-00002-of-00002.safetensors",
606
+ "vision_tower.transformer.layers.5.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
607
+ "vision_tower.transformer.layers.5.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
608
+ "vision_tower.transformer.layers.5.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
609
+ "vision_tower.transformer.layers.5.ffn_norm.weight": "model-00002-of-00002.safetensors",
610
+ "vision_tower.transformer.layers.6.attention.k_proj.weight": "model-00002-of-00002.safetensors",
611
+ "vision_tower.transformer.layers.6.attention.o_proj.weight": "model-00002-of-00002.safetensors",
612
+ "vision_tower.transformer.layers.6.attention.q_proj.weight": "model-00002-of-00002.safetensors",
613
+ "vision_tower.transformer.layers.6.attention.v_proj.weight": "model-00002-of-00002.safetensors",
614
+ "vision_tower.transformer.layers.6.attention_norm.weight": "model-00002-of-00002.safetensors",
615
+ "vision_tower.transformer.layers.6.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
616
+ "vision_tower.transformer.layers.6.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
617
+ "vision_tower.transformer.layers.6.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
618
+ "vision_tower.transformer.layers.6.ffn_norm.weight": "model-00002-of-00002.safetensors",
619
+ "vision_tower.transformer.layers.7.attention.k_proj.weight": "model-00002-of-00002.safetensors",
620
+ "vision_tower.transformer.layers.7.attention.o_proj.weight": "model-00002-of-00002.safetensors",
621
+ "vision_tower.transformer.layers.7.attention.q_proj.weight": "model-00002-of-00002.safetensors",
622
+ "vision_tower.transformer.layers.7.attention.v_proj.weight": "model-00002-of-00002.safetensors",
623
+ "vision_tower.transformer.layers.7.attention_norm.weight": "model-00002-of-00002.safetensors",
624
+ "vision_tower.transformer.layers.7.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
625
+ "vision_tower.transformer.layers.7.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
626
+ "vision_tower.transformer.layers.7.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
627
+ "vision_tower.transformer.layers.7.ffn_norm.weight": "model-00002-of-00002.safetensors",
628
+ "vision_tower.transformer.layers.8.attention.k_proj.weight": "model-00002-of-00002.safetensors",
629
+ "vision_tower.transformer.layers.8.attention.o_proj.weight": "model-00002-of-00002.safetensors",
630
+ "vision_tower.transformer.layers.8.attention.q_proj.weight": "model-00002-of-00002.safetensors",
631
+ "vision_tower.transformer.layers.8.attention.v_proj.weight": "model-00002-of-00002.safetensors",
632
+ "vision_tower.transformer.layers.8.attention_norm.weight": "model-00002-of-00002.safetensors",
633
+ "vision_tower.transformer.layers.8.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
634
+ "vision_tower.transformer.layers.8.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
635
+ "vision_tower.transformer.layers.8.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
636
+ "vision_tower.transformer.layers.8.ffn_norm.weight": "model-00002-of-00002.safetensors",
637
+ "vision_tower.transformer.layers.9.attention.k_proj.weight": "model-00002-of-00002.safetensors",
638
+ "vision_tower.transformer.layers.9.attention.o_proj.weight": "model-00002-of-00002.safetensors",
639
+ "vision_tower.transformer.layers.9.attention.q_proj.weight": "model-00002-of-00002.safetensors",
640
+ "vision_tower.transformer.layers.9.attention.v_proj.weight": "model-00002-of-00002.safetensors",
641
+ "vision_tower.transformer.layers.9.attention_norm.weight": "model-00002-of-00002.safetensors",
642
+ "vision_tower.transformer.layers.9.feed_forward.down_proj.weight": "model-00002-of-00002.safetensors",
643
+ "vision_tower.transformer.layers.9.feed_forward.gate_proj.weight": "model-00002-of-00002.safetensors",
644
+ "vision_tower.transformer.layers.9.feed_forward.up_proj.weight": "model-00002-of-00002.safetensors",
645
+ "vision_tower.transformer.layers.9.ffn_norm.weight": "model-00002-of-00002.safetensors"
646
+ }
647
+ }
params.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dim": 3072,
3
+ "n_layers": 26,
4
+ "head_dim": 128,
5
+ "hidden_dim": 9216,
6
+ "n_heads": 32,
7
+ "n_kv_heads": 8,
8
+ "rope_theta": 1000000.0,
9
+ "norm_eps": 1e-05,
10
+ "vocab_size": 131072,
11
+ "tied_embeddings": true,
12
+ "max_position_embeddings": 262144,
13
+ "llama_4_scaling": {
14
+ "original_max_position_embeddings": 16384,
15
+ "beta": 0.1
16
+ },
17
+ "q_lora_rank": null,
18
+ "qk_rope_head_dim": null,
19
+ "qk_nope_head_dim": null,
20
+ "kv_lora_rank": null,
21
+ "v_head_dim": null,
22
+ "yarn": {
23
+ "original_max_position_embeddings": 16384,
24
+ "factor": 16,
25
+ "apply_scale": false,
26
+ "beta": 32,
27
+ "alpha": 1
28
+ },
29
+ "vision_encoder": {
30
+ "image_token_id": 10,
31
+ "intermediate_size": 4096,
32
+ "num_hidden_layers": 24,
33
+ "num_attention_heads": 16,
34
+ "mm_projector_id": "patch_merge",
35
+ "spatial_merge_size": 2,
36
+ "hidden_size": 1024,
37
+ "num_channels": 3,
38
+ "image_size": 1540,
39
+ "max_image_size": 1540,
40
+ "patch_size": 14,
41
+ "rope_theta": 10000.0,
42
+ "add_pre_mm_projector_layer_norm": true,
43
+ "adapter_bias": false,
44
+ "image_break_token_id": 12,
45
+ "image_end_token_id": 13
46
+ }
47
+ }
processor_config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_break_token": "[IMG_BREAK]",
3
+ "image_end_token": "[IMG_END]",
4
+ "image_processor": {
5
+ "crop_size": null,
6
+ "data_format": "channels_first",
7
+ "device": null,
8
+ "disable_grouping": null,
9
+ "do_center_crop": null,
10
+ "do_convert_rgb": true,
11
+ "do_normalize": true,
12
+ "do_pad": null,
13
+ "do_rescale": true,
14
+ "do_resize": true,
15
+ "image_mean": [
16
+ 0.48145466,
17
+ 0.4578275,
18
+ 0.40821073
19
+ ],
20
+ "image_processor_type": "PixtralImageProcessorFast",
21
+ "image_seq_length": null,
22
+ "image_std": [
23
+ 0.26862954,
24
+ 0.26130258,
25
+ 0.27577711
26
+ ],
27
+ "input_data_format": null,
28
+ "pad_size": null,
29
+ "patch_size": 14,
30
+ "processor_class": "PixtralProcessor",
31
+ "resample": 3,
32
+ "rescale_factor": 0.00392156862745098,
33
+ "return_tensors": null,
34
+ "size": {
35
+ "longest_edge": 1540
36
+ }
37
+ },
38
+ "image_token": "[IMG]",
39
+ "patch_size": 14,
40
+ "processor_class": "PixtralProcessor",
41
+ "spatial_merge_size": 2
42
+ }
recipe.yaml ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ default_stage:
2
+ default_modifiers:
3
+ QuantizationModifier:
4
+ targets: [Linear]
5
+ ignore: ['re:.*vision_tower.*', 're:.*multi_modal_projector.*', lm_head, 're:.*embed_tokens.*']
6
+ scheme: FP8_DYNAMIC
7
+ bypass_divisibility_checks: false
8
+ requires_calibration_data: false
release-manifest.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "version": "v1",
3
+ "date": "2026-10-09",
4
+ "backbone_base": {
5
+ "repo": "StandardThinking/StandardOne-3B",
6
+ "version": "v2.2"
7
+ },
8
+ "fine_tune": {
9
+ "lora_rank": 64,
10
+ "lora_alpha": 128,
11
+ "adapter_sha256": "58a255557f5cfbf2e614bcf88e2a2f5085a308980fe6d5110f1ea82cb39fe859",
12
+ "steps": 3361,
13
+ "merge": "fp32 merge, bfloat16 save"
14
+ },
15
+ "head": {
16
+ "taps": [
17
+ "final",
18
+ 19
19
+ ],
20
+ "template": "schema-v1-nosys",
21
+ "parameters": 69513991,
22
+ "sha256": "7c7b61dd2b8bee2557b1384f08af37c0992963339c418ef467fa74fc3b3ee962"
23
+ },
24
+ "prompt_format": "chat template with an empty default system message (SH_SYSTEM_PROMPT=none)",
25
+ "release": "StandardOne-3B-SH-FP8",
26
+ "quantization": {
27
+ "source": "StandardThinking/StandardOne-3B-SH",
28
+ "source_revision": "v1",
29
+ "format": "compressed-tensors float-quantized (llm-compressor FP8_DYNAMIC layout)",
30
+ "weights": "FP8 E4M3, per-output-channel scale = absmax/448",
31
+ "activations": "FP8 E4M3 dynamic per-token (at run time)",
32
+ "quantized_tensors": 182,
33
+ "unquantized": "vision_tower, multi_modal_projector, embed_tokens, norms, lm_head",
34
+ "quantization_config_from": "StandardThinking/StandardOne-8B-FP8@4ca41ec1239c581d82fdd2f18011a205df404220",
35
+ "tool": "make_ct_fp8.py (decision-quant FP8 evaluation, not a release)"
36
+ },
37
+ "validation": {
38
+ "engine_bf16": {
39
+ "internal-eval-v1.heldout": 83.17,
40
+ "internal-eval-hardproxy-v1.heldout": 40.17,
41
+ "internal-eval-v2-realistic.heldout": 88.67,
42
+ "jevbench-easy": 100.0,
43
+ "jevbench-original": 97.22,
44
+ "jevbench-hard": 54.05,
45
+ "longctx-dev": 82.0,
46
+ "winogrande-dev": 92.69,
47
+ "repair5k-validation-policy-judgment": 58.83
48
+ },
49
+ "engine_fp8": {
50
+ "internal-eval-v1.heldout": 82.5,
51
+ "internal-eval-hardproxy-v1.heldout": 39.5,
52
+ "internal-eval-v2-realistic.heldout": 88.0,
53
+ "jevbench-easy": 100.0,
54
+ "jevbench-original": 97.22,
55
+ "jevbench-hard": 54.95,
56
+ "longctx-dev": 83.33,
57
+ "winogrande-dev": 92.39,
58
+ "repair5k-validation-policy-judgment": 57.5
59
+ },
60
+ "same_answer": [
61
+ 4681,
62
+ 4791
63
+ ],
64
+ "engine_note": "SGLang-based decision engine; ROCm native per-token FP8 quant fallback patched for padded token dims"
65
+ }
66
+ }
special_tokens_map.json ADDED
The diff for this file is too large to render. See raw diff
 
tekken.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:600bb27946565481ecf51ba8aee252e49b9a68507866080ac9c30185bb312843
3
+ size 16753784
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5f6046775b112f0e2d456ee9dba450684ab964fe5c4e231599bdc6773028135
3
+ size 17078128
tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff