abcorrea commited on
Commit
babb181
·
verified ·
1 Parent(s): 331b2d0

Upload mod2 experimental weight edit and provenance without model card

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
LICENSE ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ Apache License
3
+ Version 2.0, January 2004
4
+ http://www.apache.org/licenses/
5
+
6
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
7
+
8
+ 1. Definitions.
9
+
10
+ "License" shall mean the terms and conditions for use, reproduction,
11
+ and distribution as defined by Sections 1 through 9 of this document.
12
+
13
+ "Licensor" shall mean the copyright owner or entity authorized by
14
+ the copyright owner that is granting the License.
15
+
16
+ "Legal Entity" shall mean the union of the acting entity and all
17
+ other entities that control, are controlled by, or are under common
18
+ control with that entity. For the purposes of this definition,
19
+ "control" means (i) the power, direct or indirect, to cause the
20
+ direction or management of such entity, whether by contract or
21
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
22
+ outstanding shares, or (iii) beneficial ownership of such entity.
23
+
24
+ "You" (or "Your") shall mean an individual or Legal Entity
25
+ exercising permissions granted by this License.
26
+
27
+ "Source" form shall mean the preferred form for making modifications,
28
+ including but not limited to software source code, documentation
29
+ source, and configuration files.
30
+
31
+ "Object" form shall mean any form resulting from mechanical
32
+ transformation or translation of a Source form, including but
33
+ not limited to compiled object code, generated documentation,
34
+ and conversions to other media types.
35
+
36
+ "Work" shall mean the work of authorship, whether in Source or
37
+ Object form, made available under the License, as indicated by a
38
+ copyright notice that is included in or attached to the work
39
+ (an example is provided in the Appendix below).
40
+
41
+ "Derivative Works" shall mean any work, whether in Source or Object
42
+ form, that is based on (or derived from) the Work and for which the
43
+ editorial revisions, annotations, elaborations, or other modifications
44
+ represent, as a whole, an original work of authorship. For the purposes
45
+ of this License, Derivative Works shall not include works that remain
46
+ separable from, or merely link (or bind by name) to the interfaces of,
47
+ the Work and Derivative Works thereof.
48
+
49
+ "Contribution" shall mean any work of authorship, including
50
+ the original version of the Work and any modifications or additions
51
+ to that Work or Derivative Works thereof, that is intentionally
52
+ submitted to Licensor for inclusion in the Work by the copyright owner
53
+ or by an individual or Legal Entity authorized to submit on behalf of
54
+ the copyright owner. For the purposes of this definition, "submitted"
55
+ means any form of electronic, verbal, or written communication sent
56
+ to the Licensor or its representatives, including but not limited to
57
+ communication on electronic mailing lists, source code control systems,
58
+ and issue tracking systems that are managed by, or on behalf of, the
59
+ Licensor for the purpose of discussing and improving the Work, but
60
+ excluding communication that is conspicuously marked or otherwise
61
+ designated in writing by the copyright owner as "Not a Contribution."
62
+
63
+ "Contributor" shall mean Licensor and any individual or Legal Entity
64
+ on behalf of whom a Contribution has been received by Licensor and
65
+ subsequently incorporated within the Work.
66
+
67
+ 2. Grant of Copyright License. Subject to the terms and conditions of
68
+ this License, each Contributor hereby grants to You a perpetual,
69
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
70
+ copyright license to reproduce, prepare Derivative Works of,
71
+ publicly display, publicly perform, sublicense, and distribute the
72
+ Work and such Derivative Works in Source or Object form.
73
+
74
+ 3. Grant of Patent License. Subject to the terms and conditions of
75
+ this License, each Contributor hereby grants to You a perpetual,
76
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
77
+ (except as stated in this section) patent license to make, have made,
78
+ use, offer to sell, sell, import, and otherwise transfer the Work,
79
+ where such license applies only to those patent claims licensable
80
+ by such Contributor that are necessarily infringed by their
81
+ Contribution(s) alone or by combination of their Contribution(s)
82
+ with the Work to which such Contribution(s) was submitted. If You
83
+ institute patent litigation against any entity (including a
84
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
85
+ or a Contribution incorporated within the Work constitutes direct
86
+ or contributory patent infringement, then any patent licenses
87
+ granted to You under this License for that Work shall terminate
88
+ as of the date such litigation is filed.
89
+
90
+ 4. Redistribution. You may reproduce and distribute copies of the
91
+ Work or Derivative Works thereof in any medium, with or without
92
+ modifications, and in Source or Object form, provided that You
93
+ meet the following conditions:
94
+
95
+ (a) You must give any other recipients of the Work or
96
+ Derivative Works a copy of this License; and
97
+
98
+ (b) You must cause any modified files to carry prominent notices
99
+ stating that You changed the files; and
100
+
101
+ (c) You must retain, in the Source form of any Derivative Works
102
+ that You distribute, all copyright, patent, trademark, and
103
+ attribution notices from the Source form of the Work,
104
+ excluding those notices that do not pertain to any part of
105
+ the Derivative Works; and
106
+
107
+ (d) If the Work includes a "NOTICE" text file as part of its
108
+ distribution, then any Derivative Works that You distribute must
109
+ include a readable copy of the attribution notices contained
110
+ within such NOTICE file, excluding those notices that do not
111
+ pertain to any part of the Derivative Works, in at least one
112
+ of the following places: within a NOTICE text file distributed
113
+ as part of the Derivative Works; within the Source form or
114
+ documentation, if provided along with the Derivative Works; or,
115
+ within a display generated by the Derivative Works, if and
116
+ wherever such third-party notices normally appear. The contents
117
+ of the NOTICE file are for informational purposes only and
118
+ do not modify the License. You may add Your own attribution
119
+ notices within Derivative Works that You distribute, alongside
120
+ or as an addendum to the NOTICE text from the Work, provided
121
+ that such additional attribution notices cannot be construed
122
+ as modifying the License.
123
+
124
+ You may add Your own copyright statement to Your modifications and
125
+ may provide additional or different license terms and conditions
126
+ for use, reproduction, or distribution of Your modifications, or
127
+ for any such Derivative Works as a whole, provided Your use,
128
+ reproduction, and distribution of the Work otherwise complies with
129
+ the conditions stated in this License.
130
+
131
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
132
+ any Contribution intentionally submitted for inclusion in the Work
133
+ by You to the Licensor shall be under the terms and conditions of
134
+ this License, without any additional terms or conditions.
135
+ Notwithstanding the above, nothing herein shall supersede or modify
136
+ the terms of any separate license agreement you may have executed
137
+ with Licensor regarding such Contributions.
138
+
139
+ 6. Trademarks. This License does not grant permission to use the trade
140
+ names, trademarks, service marks, or product names of the Licensor,
141
+ except as required for reasonable and customary use in describing the
142
+ origin of the Work and reproducing the content of the NOTICE file.
143
+
144
+ 7. Disclaimer of Warranty. Unless required by applicable law or
145
+ agreed to in writing, Licensor provides the Work (and each
146
+ Contributor provides its Contributions) on an "AS IS" BASIS,
147
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
148
+ implied, including, without limitation, any warranties or conditions
149
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
150
+ PARTICULAR PURPOSE. You are solely responsible for determining the
151
+ appropriateness of using or redistributing the Work and assume any
152
+ risks associated with Your exercise of permissions under this License.
153
+
154
+ 8. Limitation of Liability. In no event and under no legal theory,
155
+ whether in tort (including negligence), contract, or otherwise,
156
+ unless required by applicable law (such as deliberate and grossly
157
+ negligent acts) or agreed to in writing, shall any Contributor be
158
+ liable to You for damages, including any direct, indirect, special,
159
+ incidental, or consequential damages of any character arising as a
160
+ result of this License or out of the use or inability to use the
161
+ Work (including but not limited to damages for loss of goodwill,
162
+ work stoppage, computer failure or malfunction, or any and all
163
+ other commercial damages or losses), even if such Contributor
164
+ has been advised of the possibility of such damages.
165
+
166
+ 9. Accepting Warranty or Additional Liability. While redistributing
167
+ the Work or Derivative Works thereof, You may choose to offer,
168
+ and charge a fee for, acceptance of support, warranty, indemnity,
169
+ or other liability obligations and/or rights consistent with this
170
+ License. However, in accepting such obligations, You may act only
171
+ on Your own behalf and on Your sole responsibility, not on behalf
172
+ of any other Contributor, and only if You agree to indemnify,
173
+ defend, and hold each Contributor harmless for any liability
174
+ incurred by, or claims asserted against, such Contributor by reason
175
+ of your accepting any such warranty or additional liability.
176
+
177
+ END OF TERMS AND CONDITIONS
178
+
179
+ APPENDIX: How to apply the Apache License to your work.
180
+
181
+ To apply the Apache License to your work, attach the following
182
+ boilerplate notice, with the fields enclosed by brackets "[]"
183
+ replaced with your own identifying information. (Don't include
184
+ the brackets!) The text should be enclosed in the appropriate
185
+ comment syntax for the file format. We also recommend that a
186
+ file or class name and description of purpose be included on the
187
+ same "printed page" as the copyright notice for easier
188
+ identification within third-party archives.
189
+
190
+ Copyright 2026 Alibaba Cloud
191
+
192
+ Licensed under the Apache License, Version 2.0 (the "License");
193
+ you may not use this file except in compliance with the License.
194
+ You may obtain a copy of the License at
195
+
196
+ http://www.apache.org/licenses/LICENSE-2.0
197
+
198
+ Unless required by applicable law or agreed to in writing, software
199
+ distributed under the License is distributed on an "AS IS" BASIS,
200
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
201
+ See the License for the specific language governing permissions and
202
+ limitations under the License.
blocksworld-999.pddl ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ (define (problem blocksworld-999)
2
+ (:domain blocksworld)
3
+ (:objects b1 b2 b3 b4 b5 b6 b7 b8 b9 b10 - object)
4
+ (:init
5
+ (arm-empty)
6
+ (clear b2)
7
+ (on b2 b10)
8
+ (on b10 b4)
9
+ (on b4 b8)
10
+ (on b8 b6)
11
+ (on b6 b3)
12
+ (on b3 b7)
13
+ (on b7 b5)
14
+ (on b5 b1)
15
+ (on-table b1)
16
+ (clear b9)
17
+ (on-table b9))
18
+ (:goal (and
19
+ (clear b3)
20
+ (on b3 b2)
21
+ (on b2 b7)
22
+ (on b7 b6)
23
+ (on b6 b10)
24
+ (on b10 b4)
25
+ (on b4 b5)
26
+ (on b5 b1)
27
+ (on b1 b8)
28
+ (on-table b8)
29
+ (clear b9)
30
+ (on-table b9))))
blocksworld-domain.pddl ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ;; Supplied by user; source:
2
+ ;; https://github.com/AI-Planning/pddl-generators/blob/main/blocksworld/domain.pddl
3
+ (define (domain blocksworld)
4
+ (:requirements :strips)
5
+ (:predicates (clear ?x) (on-table ?x) (arm-empty) (holding ?x) (on ?x ?y))
6
+ (:action pickup
7
+ :parameters (?ob)
8
+ :precondition (and (clear ?ob) (on-table ?ob) (arm-empty))
9
+ :effect (and (holding ?ob) (not (clear ?ob)) (not (on-table ?ob)) (not (arm-empty))))
10
+ (:action putdown
11
+ :parameters (?ob)
12
+ :precondition (holding ?ob)
13
+ :effect (and (clear ?ob) (arm-empty) (on-table ?ob) (not (holding ?ob))))
14
+ (:action stack
15
+ :parameters (?ob ?underob)
16
+ :precondition (and (clear ?underob) (holding ?ob))
17
+ :effect (and (arm-empty) (clear ?ob) (on ?ob ?underob)
18
+ (not (clear ?underob)) (not (holding ?ob))))
19
+ (:action unstack
20
+ :parameters (?ob ?underob)
21
+ :precondition (and (on ?ob ?underob) (clear ?ob) (arm-empty))
22
+ :effect (and (holding ?ob) (clear ?underob) (not (on ?ob ?underob))
23
+ (not (clear ?ob)) (not (arm-empty)))))
chat_template.jinja ADDED
@@ -0,0 +1,154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- set image_count = namespace(value=0) %}
2
+ {%- set video_count = namespace(value=0) %}
3
+ {%- macro render_content(content, do_vision_count, is_system_content=false) %}
4
+ {%- if content is string %}
5
+ {{- content }}
6
+ {%- elif content is iterable and content is not mapping %}
7
+ {%- for item in content %}
8
+ {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
9
+ {%- if is_system_content %}
10
+ {{- raise_exception('System message cannot contain images.') }}
11
+ {%- endif %}
12
+ {%- if do_vision_count %}
13
+ {%- set image_count.value = image_count.value + 1 %}
14
+ {%- endif %}
15
+ {%- if add_vision_id %}
16
+ {{- 'Picture ' ~ image_count.value ~ ': ' }}
17
+ {%- endif %}
18
+ {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
19
+ {%- elif 'video' in item or item.type == 'video' %}
20
+ {%- if is_system_content %}
21
+ {{- raise_exception('System message cannot contain videos.') }}
22
+ {%- endif %}
23
+ {%- if do_vision_count %}
24
+ {%- set video_count.value = video_count.value + 1 %}
25
+ {%- endif %}
26
+ {%- if add_vision_id %}
27
+ {{- 'Video ' ~ video_count.value ~ ': ' }}
28
+ {%- endif %}
29
+ {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
30
+ {%- elif 'text' in item %}
31
+ {{- item.text }}
32
+ {%- else %}
33
+ {{- raise_exception('Unexpected item type in content.') }}
34
+ {%- endif %}
35
+ {%- endfor %}
36
+ {%- elif content is none or content is undefined %}
37
+ {{- '' }}
38
+ {%- else %}
39
+ {{- raise_exception('Unexpected content type.') }}
40
+ {%- endif %}
41
+ {%- endmacro %}
42
+ {%- if not messages %}
43
+ {{- raise_exception('No messages provided.') }}
44
+ {%- endif %}
45
+ {%- if tools and tools is iterable and tools is not mapping %}
46
+ {{- '<|im_start|>system\n' }}
47
+ {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
48
+ {%- for tool in tools %}
49
+ {{- "\n" }}
50
+ {{- tool | tojson }}
51
+ {%- endfor %}
52
+ {{- "\n</tools>" }}
53
+ {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
54
+ {%- if messages[0].role == 'system' %}
55
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
56
+ {%- if content %}
57
+ {{- '\n\n' + content }}
58
+ {%- endif %}
59
+ {%- endif %}
60
+ {{- '<|im_end|>\n' }}
61
+ {%- else %}
62
+ {%- if messages[0].role == 'system' %}
63
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
64
+ {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
65
+ {%- endif %}
66
+ {%- endif %}
67
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
68
+ {%- for message in messages[::-1] %}
69
+ {%- set index = (messages|length - 1) - loop.index0 %}
70
+ {%- if ns.multi_step_tool and message.role == "user" %}
71
+ {%- set content = render_content(message.content, false)|trim %}
72
+ {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
73
+ {%- set ns.multi_step_tool = false %}
74
+ {%- set ns.last_query_index = index %}
75
+ {%- endif %}
76
+ {%- endif %}
77
+ {%- endfor %}
78
+ {%- if ns.multi_step_tool %}
79
+ {{- raise_exception('No user query found in messages.') }}
80
+ {%- endif %}
81
+ {%- for message in messages %}
82
+ {%- set content = render_content(message.content, true)|trim %}
83
+ {%- if message.role == "system" %}
84
+ {%- if not loop.first %}
85
+ {{- raise_exception('System message must be at the beginning.') }}
86
+ {%- endif %}
87
+ {%- elif message.role == "user" %}
88
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
89
+ {%- elif message.role == "assistant" %}
90
+ {%- set reasoning_content = '' %}
91
+ {%- if message.reasoning_content is string %}
92
+ {%- set reasoning_content = message.reasoning_content %}
93
+ {%- else %}
94
+ {%- if '</think>' in content %}
95
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
96
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
97
+ {%- endif %}
98
+ {%- endif %}
99
+ {%- set reasoning_content = reasoning_content|trim %}
100
+ {%- if loop.index0 > ns.last_query_index %}
101
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
102
+ {%- else %}
103
+ {{- '<|im_start|>' + message.role + '\n' + content }}
104
+ {%- endif %}
105
+ {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
106
+ {%- for tool_call in message.tool_calls %}
107
+ {%- if tool_call.function is defined %}
108
+ {%- set tool_call = tool_call.function %}
109
+ {%- endif %}
110
+ {%- if loop.first %}
111
+ {%- if content|trim %}
112
+ {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
113
+ {%- else %}
114
+ {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
115
+ {%- endif %}
116
+ {%- else %}
117
+ {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
118
+ {%- endif %}
119
+ {%- if tool_call.arguments is defined %}
120
+ {%- for args_name, args_value in tool_call.arguments|items %}
121
+ {{- '<parameter=' + args_name + '>\n' }}
122
+ {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
123
+ {{- args_value }}
124
+ {{- '\n</parameter>\n' }}
125
+ {%- endfor %}
126
+ {%- endif %}
127
+ {{- '</function>\n</tool_call>' }}
128
+ {%- endfor %}
129
+ {%- endif %}
130
+ {{- '<|im_end|>\n' }}
131
+ {%- elif message.role == "tool" %}
132
+ {%- if loop.previtem and loop.previtem.role != "tool" %}
133
+ {{- '<|im_start|>user' }}
134
+ {%- endif %}
135
+ {{- '\n<tool_response>\n' }}
136
+ {{- content }}
137
+ {{- '\n</tool_response>' }}
138
+ {%- if not loop.last and loop.nextitem.role != "tool" %}
139
+ {{- '<|im_end|>\n' }}
140
+ {%- elif loop.last %}
141
+ {{- '<|im_end|>\n' }}
142
+ {%- endif %}
143
+ {%- else %}
144
+ {{- raise_exception('Unexpected message role.') }}
145
+ {%- endif %}
146
+ {%- endfor %}
147
+ {%- if add_generation_prompt %}
148
+ {{- '<|im_start|>assistant\n' }}
149
+ {%- if enable_thinking is defined and enable_thinking is true %}
150
+ {{- '<think>\n' }}
151
+ {%- else %}
152
+ {{- '<think>\n\n</think>\n\n' }}
153
+ {%- endif %}
154
+ {%- endif %}
config.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3_5ForConditionalGeneration"
4
+ ],
5
+ "image_token_id": 248056,
6
+ "model_type": "qwen3_5",
7
+ "text_config": {
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "attn_output_gate": true,
11
+ "dtype": "bfloat16",
12
+ "eos_token_id": 248044,
13
+ "full_attention_interval": 4,
14
+ "head_dim": 256,
15
+ "hidden_act": "silu",
16
+ "hidden_size": 1024,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 3584,
19
+ "layer_types": [
20
+ "linear_attention",
21
+ "linear_attention",
22
+ "linear_attention",
23
+ "full_attention",
24
+ "linear_attention",
25
+ "linear_attention",
26
+ "linear_attention",
27
+ "full_attention",
28
+ "linear_attention",
29
+ "linear_attention",
30
+ "linear_attention",
31
+ "full_attention",
32
+ "linear_attention",
33
+ "linear_attention",
34
+ "linear_attention",
35
+ "full_attention",
36
+ "linear_attention",
37
+ "linear_attention",
38
+ "linear_attention",
39
+ "full_attention",
40
+ "linear_attention",
41
+ "linear_attention",
42
+ "linear_attention",
43
+ "full_attention"
44
+ ],
45
+ "linear_conv_kernel_dim": 4,
46
+ "linear_key_head_dim": 128,
47
+ "linear_num_key_heads": 16,
48
+ "linear_num_value_heads": 16,
49
+ "linear_value_head_dim": 128,
50
+ "max_position_embeddings": 262144,
51
+ "mlp_only_layers": [],
52
+ "model_type": "qwen3_5_text",
53
+ "mtp_num_hidden_layers": 1,
54
+ "mtp_use_dedicated_embeddings": false,
55
+ "num_attention_heads": 8,
56
+ "num_hidden_layers": 24,
57
+ "num_key_value_heads": 2,
58
+ "rms_norm_eps": 1e-06,
59
+ "tie_word_embeddings": true,
60
+ "use_cache": true,
61
+ "vocab_size": 248320,
62
+ "mamba_ssm_dtype": "float32",
63
+ "rope_parameters": {
64
+ "mrope_interleaved": true,
65
+ "mrope_section": [
66
+ 11,
67
+ 11,
68
+ 10
69
+ ],
70
+ "rope_type": "default",
71
+ "rope_theta": 10000000,
72
+ "partial_rotary_factor": 0.25
73
+ }
74
+ },
75
+ "tie_word_embeddings": true,
76
+ "transformers_version": "4.57.0.dev0",
77
+ "video_token_id": 248057,
78
+ "vision_config": {
79
+ "deepstack_visual_indexes": [],
80
+ "depth": 12,
81
+ "hidden_act": "gelu_pytorch_tanh",
82
+ "hidden_size": 768,
83
+ "in_channels": 3,
84
+ "initializer_range": 0.02,
85
+ "intermediate_size": 3072,
86
+ "model_type": "qwen3_5",
87
+ "num_heads": 12,
88
+ "num_position_embeddings": 2304,
89
+ "out_hidden_size": 1024,
90
+ "patch_size": 16,
91
+ "spatial_merge_size": 2,
92
+ "temporal_patch_size": 2
93
+ },
94
+ "vision_end_token_id": 248054,
95
+ "vision_start_token_id": 248053
96
+ }
download_manifest.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "repo": "Qwen/Qwen3.5-0.8B",
3
+ "revision": "2fc06364715b967f1860aea9cf38778875588b17",
4
+ "files": {
5
+ ".gitattributes": {
6
+ "sha256": "34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930",
7
+ "bytes": 1570
8
+ },
9
+ "LICENSE": {
10
+ "sha256": "bbedc3fda3305820b977265f01b8619d87570a6739de3a5582c3464840f1e57a",
11
+ "bytes": 11544
12
+ },
13
+ "README.md": {
14
+ "sha256": "87a163af54f32fa608a0f8d3ac67945c53dd2b4a7c96740b3d7fdc28e8458864",
15
+ "bytes": 61705
16
+ },
17
+ "chat_template.jinja": {
18
+ "sha256": "273d8e0e683b885071fb17e08d71e5f2a5ddfb5309756181681de4f5a1822d80",
19
+ "bytes": 7755
20
+ },
21
+ "config.json": {
22
+ "sha256": "b90b86f35c8e6925ef74ee04d0e758f0a845c83a42089ad82bbaa948de9b4204",
23
+ "bytes": 2907
24
+ },
25
+ "merges.txt": {
26
+ "sha256": "a9d356d7bdf1ef4949e3e748e95b8e10ad9d4e2e838eddc38a0a7b6b94d1db8d",
27
+ "bytes": 3353259
28
+ },
29
+ "model.safetensors-00001-of-00001.safetensors": {
30
+ "sha256": "04b1c301231dd422b8860db31311ab2721511346a32cb1e079c4c4e5f1fe4696",
31
+ "bytes": 1746942600
32
+ },
33
+ "model.safetensors.index.json": {
34
+ "sha256": "d8a08838a613b025eb7952ed9db11696213e57e76a375661ef5c12f9dd5dcf4e",
35
+ "bytes": 50900
36
+ },
37
+ "preprocessor_config.json": {
38
+ "sha256": "27225450ac9c6529872ee1924fcb0962ff5634834f817040f444118116f4e516",
39
+ "bytes": 390
40
+ },
41
+ "tokenizer.json": {
42
+ "sha256": "5f9e4d4901a92b997e463c1f46055088b6cca5ca61a6522d1b9f64c4bb81cb42",
43
+ "bytes": 12807982
44
+ },
45
+ "tokenizer_config.json": {
46
+ "sha256": "49e2b6e395f959f077f1e992b338919c0d4a9732fc6e613995e06557f843500c",
47
+ "bytes": 16709
48
+ },
49
+ "video_preprocessor_config.json": {
50
+ "sha256": "7768af27c1fafa9cc9011c1dc20067e03f8915e03b63504550e11d5066986d13",
51
+ "bytes": 385
52
+ },
53
+ "vocab.json": {
54
+ "sha256": "ce99b4cb2983d118806ce0a8b777a35b093e2000a503ebde25853284c9dfa003",
55
+ "bytes": 6722759
56
+ }
57
+ }
58
+ }
edit_report.json ADDED
@@ -0,0 +1,1313 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "version": 2,
3
+ "status": "experimental_unvalidated",
4
+ "improvement_established": false,
5
+ "forward_passes": 0,
6
+ "neural_network_runtime_imported": false,
7
+ "base_model": "Qwen/Qwen3.5-0.8B",
8
+ "base_revision": "2fc06364715b967f1860aea9cf38778875588b17",
9
+ "parent_checkpoint": "Qwen3.5-0.8B-blocksworld-experimental",
10
+ "parent_sha256": "12d6e276309afe792154864b2f96dd23ed6b75d8f403d6dd9029a5388901e123",
11
+ "method": "Incremental 1.01x MLP output-column scaling using 17 separate static PDDL signals",
12
+ "selection": {
13
+ "layer_range_inclusive": [
14
+ 12,
15
+ 23
16
+ ],
17
+ "scale": 1.01,
18
+ "columns": 68,
19
+ "per_direction": 4,
20
+ "max_per_layer": 8
21
+ },
22
+ "exact_symbols": [
23
+ "clear",
24
+ "on-table",
25
+ "arm-empty",
26
+ "holding",
27
+ "on",
28
+ "pickup",
29
+ "putdown",
30
+ "stack",
31
+ "unstack"
32
+ ],
33
+ "tokenization": {
34
+ "clear": [
35
+ {
36
+ "context": "clear",
37
+ "span": [
38
+ 0,
39
+ 5
40
+ ],
41
+ "ids": [
42
+ 7187
43
+ ],
44
+ "tokens": [
45
+ "clear"
46
+ ]
47
+ },
48
+ {
49
+ "context": "(clear ?x)",
50
+ "span": [
51
+ 1,
52
+ 6
53
+ ],
54
+ "ids": [
55
+ 1290,
56
+ 1493
57
+ ],
58
+ "tokens": [
59
+ "(c",
60
+ "lear"
61
+ ]
62
+ },
63
+ {
64
+ "context": " (clear ?x)",
65
+ "span": [
66
+ 5,
67
+ 10
68
+ ],
69
+ "ids": [
70
+ 7187
71
+ ],
72
+ "tokens": [
73
+ "clear"
74
+ ]
75
+ }
76
+ ],
77
+ "on-table": [
78
+ {
79
+ "context": "on-table",
80
+ "span": [
81
+ 0,
82
+ 8
83
+ ],
84
+ "ids": [
85
+ 263,
86
+ 16293
87
+ ],
88
+ "tokens": [
89
+ "on",
90
+ "-table"
91
+ ]
92
+ },
93
+ {
94
+ "context": "(on-table ?x)",
95
+ "span": [
96
+ 1,
97
+ 9
98
+ ],
99
+ "ids": [
100
+ 28395,
101
+ 16293
102
+ ],
103
+ "tokens": [
104
+ "(on",
105
+ "-table"
106
+ ]
107
+ },
108
+ {
109
+ "context": " (on-table ?x)",
110
+ "span": [
111
+ 5,
112
+ 13
113
+ ],
114
+ "ids": [
115
+ 263,
116
+ 16293
117
+ ],
118
+ "tokens": [
119
+ "on",
120
+ "-table"
121
+ ]
122
+ }
123
+ ],
124
+ "arm-empty": [
125
+ {
126
+ "context": "arm-empty",
127
+ "span": [
128
+ 0,
129
+ 9
130
+ ],
131
+ "ids": [
132
+ 2109,
133
+ 38127
134
+ ],
135
+ "tokens": [
136
+ "arm",
137
+ "-empty"
138
+ ]
139
+ },
140
+ {
141
+ "context": "(arm-empty)",
142
+ "span": [
143
+ 1,
144
+ 10
145
+ ],
146
+ "ids": [
147
+ 2109,
148
+ 38127
149
+ ],
150
+ "tokens": [
151
+ "arm",
152
+ "-empty"
153
+ ]
154
+ },
155
+ {
156
+ "context": " (arm-empty)",
157
+ "span": [
158
+ 5,
159
+ 14
160
+ ],
161
+ "ids": [
162
+ 2109,
163
+ 38127
164
+ ],
165
+ "tokens": [
166
+ "arm",
167
+ "-empty"
168
+ ]
169
+ }
170
+ ],
171
+ "holding": [
172
+ {
173
+ "context": "holding",
174
+ "span": [
175
+ 0,
176
+ 7
177
+ ],
178
+ "ids": [
179
+ 41078
180
+ ],
181
+ "tokens": [
182
+ "holding"
183
+ ]
184
+ },
185
+ {
186
+ "context": "(holding ?x)",
187
+ "span": [
188
+ 1,
189
+ 8
190
+ ],
191
+ "ids": [
192
+ 3101,
193
+ 14561
194
+ ],
195
+ "tokens": [
196
+ "(h",
197
+ "olding"
198
+ ]
199
+ },
200
+ {
201
+ "context": " (holding ?x)",
202
+ "span": [
203
+ 5,
204
+ 12
205
+ ],
206
+ "ids": [
207
+ 41078
208
+ ],
209
+ "tokens": [
210
+ "holding"
211
+ ]
212
+ }
213
+ ],
214
+ "on": [
215
+ {
216
+ "context": "on",
217
+ "span": [
218
+ 0,
219
+ 2
220
+ ],
221
+ "ids": [
222
+ 263
223
+ ],
224
+ "tokens": [
225
+ "on"
226
+ ]
227
+ },
228
+ {
229
+ "context": "(on ?x ?y)",
230
+ "span": [
231
+ 1,
232
+ 3
233
+ ],
234
+ "ids": [
235
+ 28395
236
+ ],
237
+ "tokens": [
238
+ "(on"
239
+ ]
240
+ },
241
+ {
242
+ "context": " (on ?x ?y)",
243
+ "span": [
244
+ 5,
245
+ 7
246
+ ],
247
+ "ids": [
248
+ 263
249
+ ],
250
+ "tokens": [
251
+ "on"
252
+ ]
253
+ }
254
+ ],
255
+ "pickup": [
256
+ {
257
+ "context": "pickup",
258
+ "span": [
259
+ 0,
260
+ 6
261
+ ],
262
+ "ids": [
263
+ 81762
264
+ ],
265
+ "tokens": [
266
+ "pickup"
267
+ ]
268
+ },
269
+ {
270
+ "context": "(:action pickup\n",
271
+ "span": [
272
+ 9,
273
+ 15
274
+ ],
275
+ "ids": [
276
+ 28619
277
+ ],
278
+ "tokens": [
279
+ "\u0120pickup"
280
+ ]
281
+ },
282
+ {
283
+ "context": "(pickup ?ob)",
284
+ "span": [
285
+ 1,
286
+ 7
287
+ ],
288
+ "ids": [
289
+ 1250,
290
+ 839,
291
+ 446
292
+ ],
293
+ "tokens": [
294
+ "(p",
295
+ "ick",
296
+ "up"
297
+ ]
298
+ }
299
+ ],
300
+ "putdown": [
301
+ {
302
+ "context": "putdown",
303
+ "span": [
304
+ 0,
305
+ 7
306
+ ],
307
+ "ids": [
308
+ 611,
309
+ 2829
310
+ ],
311
+ "tokens": [
312
+ "put",
313
+ "down"
314
+ ]
315
+ },
316
+ {
317
+ "context": "(:action putdown\n",
318
+ "span": [
319
+ 9,
320
+ 16
321
+ ],
322
+ "ids": [
323
+ 2113,
324
+ 2829
325
+ ],
326
+ "tokens": [
327
+ "\u0120put",
328
+ "down"
329
+ ]
330
+ },
331
+ {
332
+ "context": "(putdown ?ob)",
333
+ "span": [
334
+ 1,
335
+ 8
336
+ ],
337
+ "ids": [
338
+ 611,
339
+ 2829
340
+ ],
341
+ "tokens": [
342
+ "put",
343
+ "down"
344
+ ]
345
+ }
346
+ ],
347
+ "stack": [
348
+ {
349
+ "context": "stack",
350
+ "span": [
351
+ 0,
352
+ 5
353
+ ],
354
+ "ids": [
355
+ 7450
356
+ ],
357
+ "tokens": [
358
+ "stack"
359
+ ]
360
+ },
361
+ {
362
+ "context": "(:action stack\n",
363
+ "span": [
364
+ 9,
365
+ 14
366
+ ],
367
+ "ids": [
368
+ 5436
369
+ ],
370
+ "tokens": [
371
+ "\u0120stack"
372
+ ]
373
+ },
374
+ {
375
+ "context": "(stack ?ob)",
376
+ "span": [
377
+ 1,
378
+ 6
379
+ ],
380
+ "ids": [
381
+ 26549
382
+ ],
383
+ "tokens": [
384
+ "(stack"
385
+ ]
386
+ }
387
+ ],
388
+ "unstack": [
389
+ {
390
+ "context": "unstack",
391
+ "span": [
392
+ 0,
393
+ 7
394
+ ],
395
+ "ids": [
396
+ 354,
397
+ 7450
398
+ ],
399
+ "tokens": [
400
+ "un",
401
+ "stack"
402
+ ]
403
+ },
404
+ {
405
+ "context": "(:action unstack\n",
406
+ "span": [
407
+ 9,
408
+ 16
409
+ ],
410
+ "ids": [
411
+ 632,
412
+ 7450
413
+ ],
414
+ "tokens": [
415
+ "\u0120un",
416
+ "stack"
417
+ ]
418
+ },
419
+ {
420
+ "context": "(unstack ?ob)",
421
+ "span": [
422
+ 1,
423
+ 8
424
+ ],
425
+ "ids": [
426
+ 17818,
427
+ 7450
428
+ ],
429
+ "tokens": [
430
+ "(un",
431
+ "stack"
432
+ ]
433
+ }
434
+ ]
435
+ },
436
+ "direction_definitions": {
437
+ "symbol": "unit(symbol vector - centroid of the 9 exact domain symbols)",
438
+ "preconditions": "unit(mean vectors of precondition predicate names - centroid of the 5 predicates)",
439
+ "effects": "unit(mean vectors of add-effect predicate names - mean vectors of delete-effect predicate names)",
440
+ "symbol_vector": "Mean of normalized token weight rows overlapping the identifier in 3 lexical/PDDL contexts; normalize each context and then their mean",
441
+ "selection": "Round-robin top positive cosine match, 4 unique columns per signal, at most 8 per layer"
442
+ },
443
+ "domain_sha256": "191f3ff3cfe966ca8d277dd7f5269543cd19d0a6461e73b7e8b4507a09c8e5e1",
444
+ "example_sha256": "7303c346f9182d57372190bb0fc0d86d382025ae217c7c42f1af6b6d29d6fd81",
445
+ "symbolic_checks": {
446
+ "rules": {
447
+ "reachable_three_block_states": 22,
448
+ "legal_transitions_checked": 42,
449
+ "state_invariants_passed": true,
450
+ "inverse_actions_passed": true
451
+ },
452
+ "example": {
453
+ "goal_satisfied_symbolically": true,
454
+ "plan_length": 32,
455
+ "plan": [
456
+ [
457
+ "unstack",
458
+ "b2",
459
+ "b10"
460
+ ],
461
+ [
462
+ "putdown",
463
+ "b2"
464
+ ],
465
+ [
466
+ "unstack",
467
+ "b10",
468
+ "b4"
469
+ ],
470
+ [
471
+ "putdown",
472
+ "b10"
473
+ ],
474
+ [
475
+ "unstack",
476
+ "b4",
477
+ "b8"
478
+ ],
479
+ [
480
+ "putdown",
481
+ "b4"
482
+ ],
483
+ [
484
+ "unstack",
485
+ "b8",
486
+ "b6"
487
+ ],
488
+ [
489
+ "putdown",
490
+ "b8"
491
+ ],
492
+ [
493
+ "unstack",
494
+ "b6",
495
+ "b3"
496
+ ],
497
+ [
498
+ "putdown",
499
+ "b6"
500
+ ],
501
+ [
502
+ "unstack",
503
+ "b3",
504
+ "b7"
505
+ ],
506
+ [
507
+ "putdown",
508
+ "b3"
509
+ ],
510
+ [
511
+ "unstack",
512
+ "b7",
513
+ "b5"
514
+ ],
515
+ [
516
+ "putdown",
517
+ "b7"
518
+ ],
519
+ [
520
+ "unstack",
521
+ "b5",
522
+ "b1"
523
+ ],
524
+ [
525
+ "putdown",
526
+ "b5"
527
+ ],
528
+ [
529
+ "pickup",
530
+ "b1"
531
+ ],
532
+ [
533
+ "stack",
534
+ "b1",
535
+ "b8"
536
+ ],
537
+ [
538
+ "pickup",
539
+ "b5"
540
+ ],
541
+ [
542
+ "stack",
543
+ "b5",
544
+ "b1"
545
+ ],
546
+ [
547
+ "pickup",
548
+ "b4"
549
+ ],
550
+ [
551
+ "stack",
552
+ "b4",
553
+ "b5"
554
+ ],
555
+ [
556
+ "pickup",
557
+ "b10"
558
+ ],
559
+ [
560
+ "stack",
561
+ "b10",
562
+ "b4"
563
+ ],
564
+ [
565
+ "pickup",
566
+ "b6"
567
+ ],
568
+ [
569
+ "stack",
570
+ "b6",
571
+ "b10"
572
+ ],
573
+ [
574
+ "pickup",
575
+ "b7"
576
+ ],
577
+ [
578
+ "stack",
579
+ "b7",
580
+ "b6"
581
+ ],
582
+ [
583
+ "pickup",
584
+ "b2"
585
+ ],
586
+ [
587
+ "stack",
588
+ "b2",
589
+ "b7"
590
+ ],
591
+ [
592
+ "pickup",
593
+ "b3"
594
+ ],
595
+ [
596
+ "stack",
597
+ "b3",
598
+ "b2"
599
+ ]
600
+ ],
601
+ "model_evaluation": false,
602
+ "optimality_claimed": false
603
+ }
604
+ },
605
+ "symbolic_check_scope": "Checks validate the PDDL interpretation only; the model produced no plan and was not evaluated.",
606
+ "edits": [
607
+ {
608
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
609
+ "layer": 18,
610
+ "column": 2263,
611
+ "static_cosine": 0.4291132688522339,
612
+ "selection_signal": "symbol:clear",
613
+ "also_edited_in_v1": false,
614
+ "changed_scalars": 1024,
615
+ "relative_l2_change": 0.009879549033939838
616
+ },
617
+ {
618
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
619
+ "layer": 23,
620
+ "column": 419,
621
+ "static_cosine": 0.3148460388183594,
622
+ "selection_signal": "symbol:on-table",
623
+ "also_edited_in_v1": false,
624
+ "changed_scalars": 1024,
625
+ "relative_l2_change": 0.010126868262887001
626
+ },
627
+ {
628
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
629
+ "layer": 23,
630
+ "column": 2583,
631
+ "static_cosine": 0.20350371301174164,
632
+ "selection_signal": "symbol:arm-empty",
633
+ "also_edited_in_v1": false,
634
+ "changed_scalars": 1024,
635
+ "relative_l2_change": 0.009840910322964191
636
+ },
637
+ {
638
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
639
+ "layer": 23,
640
+ "column": 2788,
641
+ "static_cosine": 0.43104276061058044,
642
+ "selection_signal": "symbol:holding",
643
+ "also_edited_in_v1": false,
644
+ "changed_scalars": 1024,
645
+ "relative_l2_change": 0.010035578161478043
646
+ },
647
+ {
648
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
649
+ "layer": 23,
650
+ "column": 1667,
651
+ "static_cosine": 0.4818803668022156,
652
+ "selection_signal": "symbol:on",
653
+ "also_edited_in_v1": false,
654
+ "changed_scalars": 1024,
655
+ "relative_l2_change": 0.009887542575597763
656
+ },
657
+ {
658
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
659
+ "layer": 23,
660
+ "column": 806,
661
+ "static_cosine": 0.22057151794433594,
662
+ "selection_signal": "symbol:pickup",
663
+ "also_edited_in_v1": false,
664
+ "changed_scalars": 1024,
665
+ "relative_l2_change": 0.010023032315075397
666
+ },
667
+ {
668
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
669
+ "layer": 20,
670
+ "column": 2061,
671
+ "static_cosine": 0.32859548926353455,
672
+ "selection_signal": "symbol:putdown",
673
+ "also_edited_in_v1": false,
674
+ "changed_scalars": 1024,
675
+ "relative_l2_change": 0.009840171784162521
676
+ },
677
+ {
678
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
679
+ "layer": 18,
680
+ "column": 2217,
681
+ "static_cosine": 0.31344717741012573,
682
+ "selection_signal": "symbol:stack",
683
+ "also_edited_in_v1": false,
684
+ "changed_scalars": 1024,
685
+ "relative_l2_change": 0.009877270087599754
686
+ },
687
+ {
688
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
689
+ "layer": 23,
690
+ "column": 1174,
691
+ "static_cosine": 0.39072591066360474,
692
+ "selection_signal": "symbol:unstack",
693
+ "also_edited_in_v1": false,
694
+ "changed_scalars": 1024,
695
+ "relative_l2_change": 0.009870754554867744
696
+ },
697
+ {
698
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
699
+ "layer": 20,
700
+ "column": 2907,
701
+ "static_cosine": 0.2374749630689621,
702
+ "selection_signal": "preconditions:pickup",
703
+ "also_edited_in_v1": false,
704
+ "changed_scalars": 1024,
705
+ "relative_l2_change": 0.009936168789863586
706
+ },
707
+ {
708
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
709
+ "layer": 21,
710
+ "column": 846,
711
+ "static_cosine": 0.349545419216156,
712
+ "selection_signal": "effects:pickup",
713
+ "also_edited_in_v1": false,
714
+ "changed_scalars": 1024,
715
+ "relative_l2_change": 0.009844872169196606
716
+ },
717
+ {
718
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
719
+ "layer": 20,
720
+ "column": 2393,
721
+ "static_cosine": 0.30698007345199585,
722
+ "selection_signal": "preconditions:putdown",
723
+ "also_edited_in_v1": false,
724
+ "changed_scalars": 1024,
725
+ "relative_l2_change": 0.009781492874026299
726
+ },
727
+ {
728
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
729
+ "layer": 22,
730
+ "column": 3003,
731
+ "static_cosine": 0.25874191522598267,
732
+ "selection_signal": "effects:putdown",
733
+ "also_edited_in_v1": false,
734
+ "changed_scalars": 1024,
735
+ "relative_l2_change": 0.01003392692655325
736
+ },
737
+ {
738
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
739
+ "layer": 22,
740
+ "column": 882,
741
+ "static_cosine": 0.30257508158683777,
742
+ "selection_signal": "preconditions:stack",
743
+ "also_edited_in_v1": false,
744
+ "changed_scalars": 1024,
745
+ "relative_l2_change": 0.009734613820910454
746
+ },
747
+ {
748
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
749
+ "layer": 22,
750
+ "column": 972,
751
+ "static_cosine": 0.2577032744884491,
752
+ "selection_signal": "effects:stack",
753
+ "also_edited_in_v1": false,
754
+ "changed_scalars": 1024,
755
+ "relative_l2_change": 0.009932386688888073
756
+ },
757
+ {
758
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
759
+ "layer": 23,
760
+ "column": 781,
761
+ "static_cosine": 0.3023082911968231,
762
+ "selection_signal": "preconditions:unstack",
763
+ "also_edited_in_v1": false,
764
+ "changed_scalars": 1024,
765
+ "relative_l2_change": 0.009753502905368805
766
+ },
767
+ {
768
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
769
+ "layer": 19,
770
+ "column": 1498,
771
+ "static_cosine": 0.22243927419185638,
772
+ "selection_signal": "effects:unstack",
773
+ "also_edited_in_v1": false,
774
+ "changed_scalars": 1024,
775
+ "relative_l2_change": 0.009854796342551708
776
+ },
777
+ {
778
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
779
+ "layer": 19,
780
+ "column": 1046,
781
+ "static_cosine": 0.4281984865665436,
782
+ "selection_signal": "symbol:clear",
783
+ "also_edited_in_v1": false,
784
+ "changed_scalars": 1024,
785
+ "relative_l2_change": 0.010013757273554802
786
+ },
787
+ {
788
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
789
+ "layer": 22,
790
+ "column": 1672,
791
+ "static_cosine": 0.25062596797943115,
792
+ "selection_signal": "symbol:on-table",
793
+ "also_edited_in_v1": false,
794
+ "changed_scalars": 1024,
795
+ "relative_l2_change": 0.00967103149741888
796
+ },
797
+ {
798
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
799
+ "layer": 23,
800
+ "column": 3152,
801
+ "static_cosine": 0.1842198669910431,
802
+ "selection_signal": "symbol:arm-empty",
803
+ "also_edited_in_v1": false,
804
+ "changed_scalars": 1024,
805
+ "relative_l2_change": 0.009924721904098988
806
+ },
807
+ {
808
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
809
+ "layer": 19,
810
+ "column": 380,
811
+ "static_cosine": 0.26562824845314026,
812
+ "selection_signal": "symbol:holding",
813
+ "also_edited_in_v1": false,
814
+ "changed_scalars": 1024,
815
+ "relative_l2_change": 0.009910458698868752
816
+ },
817
+ {
818
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
819
+ "layer": 21,
820
+ "column": 2407,
821
+ "static_cosine": 0.42676302790641785,
822
+ "selection_signal": "symbol:on",
823
+ "also_edited_in_v1": false,
824
+ "changed_scalars": 1024,
825
+ "relative_l2_change": 0.009762873873114586
826
+ },
827
+ {
828
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
829
+ "layer": 21,
830
+ "column": 1091,
831
+ "static_cosine": 0.16545610129833221,
832
+ "selection_signal": "symbol:pickup",
833
+ "also_edited_in_v1": false,
834
+ "changed_scalars": 1024,
835
+ "relative_l2_change": 0.009957507252693176
836
+ },
837
+ {
838
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
839
+ "layer": 22,
840
+ "column": 1963,
841
+ "static_cosine": 0.30681341886520386,
842
+ "selection_signal": "symbol:putdown",
843
+ "also_edited_in_v1": false,
844
+ "changed_scalars": 1024,
845
+ "relative_l2_change": 0.00997726246714592
846
+ },
847
+ {
848
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
849
+ "layer": 21,
850
+ "column": 547,
851
+ "static_cosine": 0.1895275115966797,
852
+ "selection_signal": "symbol:stack",
853
+ "also_edited_in_v1": false,
854
+ "changed_scalars": 1024,
855
+ "relative_l2_change": 0.010087314993143082
856
+ },
857
+ {
858
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
859
+ "layer": 21,
860
+ "column": 1255,
861
+ "static_cosine": 0.3766050338745117,
862
+ "selection_signal": "symbol:unstack",
863
+ "also_edited_in_v1": false,
864
+ "changed_scalars": 1024,
865
+ "relative_l2_change": 0.010032253339886665
866
+ },
867
+ {
868
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
869
+ "layer": 18,
870
+ "column": 913,
871
+ "static_cosine": 0.21852751076221466,
872
+ "selection_signal": "preconditions:pickup",
873
+ "also_edited_in_v1": false,
874
+ "changed_scalars": 1024,
875
+ "relative_l2_change": 0.009869366884231567
876
+ },
877
+ {
878
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
879
+ "layer": 20,
880
+ "column": 3185,
881
+ "static_cosine": 0.26854634284973145,
882
+ "selection_signal": "effects:pickup",
883
+ "also_edited_in_v1": false,
884
+ "changed_scalars": 1024,
885
+ "relative_l2_change": 0.010148510336875916
886
+ },
887
+ {
888
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
889
+ "layer": 17,
890
+ "column": 886,
891
+ "static_cosine": 0.24109981954097748,
892
+ "selection_signal": "preconditions:putdown",
893
+ "also_edited_in_v1": false,
894
+ "changed_scalars": 1024,
895
+ "relative_l2_change": 0.009785356000065804
896
+ },
897
+ {
898
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
899
+ "layer": 19,
900
+ "column": 3405,
901
+ "static_cosine": 0.2075931429862976,
902
+ "selection_signal": "effects:putdown",
903
+ "also_edited_in_v1": false,
904
+ "changed_scalars": 1024,
905
+ "relative_l2_change": 0.010182139463722706
906
+ },
907
+ {
908
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
909
+ "layer": 18,
910
+ "column": 2495,
911
+ "static_cosine": 0.2720789611339569,
912
+ "selection_signal": "preconditions:stack",
913
+ "also_edited_in_v1": false,
914
+ "changed_scalars": 1024,
915
+ "relative_l2_change": 0.009848741814494133
916
+ },
917
+ {
918
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
919
+ "layer": 20,
920
+ "column": 101,
921
+ "static_cosine": 0.22035039961338043,
922
+ "selection_signal": "effects:stack",
923
+ "also_edited_in_v1": false,
924
+ "changed_scalars": 1024,
925
+ "relative_l2_change": 0.010001313872635365
926
+ },
927
+ {
928
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
929
+ "layer": 19,
930
+ "column": 2369,
931
+ "static_cosine": 0.27164438366889954,
932
+ "selection_signal": "preconditions:unstack",
933
+ "also_edited_in_v1": false,
934
+ "changed_scalars": 1024,
935
+ "relative_l2_change": 0.01007683202624321
936
+ },
937
+ {
938
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
939
+ "layer": 19,
940
+ "column": 1932,
941
+ "static_cosine": 0.20632174611091614,
942
+ "selection_signal": "effects:unstack",
943
+ "also_edited_in_v1": false,
944
+ "changed_scalars": 1024,
945
+ "relative_l2_change": 0.009952910244464874
946
+ },
947
+ {
948
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
949
+ "layer": 17,
950
+ "column": 2887,
951
+ "static_cosine": 0.1427300125360489,
952
+ "selection_signal": "symbol:clear",
953
+ "also_edited_in_v1": false,
954
+ "changed_scalars": 1024,
955
+ "relative_l2_change": 0.010004010051488876
956
+ },
957
+ {
958
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
959
+ "layer": 20,
960
+ "column": 2019,
961
+ "static_cosine": 0.24661760032176971,
962
+ "selection_signal": "symbol:on-table",
963
+ "also_edited_in_v1": false,
964
+ "changed_scalars": 1024,
965
+ "relative_l2_change": 0.009791170246899128
966
+ },
967
+ {
968
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
969
+ "layer": 22,
970
+ "column": 597,
971
+ "static_cosine": 0.1726370006799698,
972
+ "selection_signal": "symbol:arm-empty",
973
+ "also_edited_in_v1": false,
974
+ "changed_scalars": 1024,
975
+ "relative_l2_change": 0.009881876409053802
976
+ },
977
+ {
978
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
979
+ "layer": 18,
980
+ "column": 3557,
981
+ "static_cosine": 0.23936277627944946,
982
+ "selection_signal": "symbol:holding",
983
+ "also_edited_in_v1": false,
984
+ "changed_scalars": 1024,
985
+ "relative_l2_change": 0.009727371856570244
986
+ },
987
+ {
988
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
989
+ "layer": 21,
990
+ "column": 1094,
991
+ "static_cosine": 0.398653507232666,
992
+ "selection_signal": "symbol:on",
993
+ "also_edited_in_v1": false,
994
+ "changed_scalars": 1024,
995
+ "relative_l2_change": 0.010150876827538013
996
+ },
997
+ {
998
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
999
+ "layer": 20,
1000
+ "column": 3553,
1001
+ "static_cosine": 0.16080619394779205,
1002
+ "selection_signal": "symbol:pickup",
1003
+ "also_edited_in_v1": false,
1004
+ "changed_scalars": 1024,
1005
+ "relative_l2_change": 0.010120393708348274
1006
+ },
1007
+ {
1008
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1009
+ "layer": 19,
1010
+ "column": 1465,
1011
+ "static_cosine": 0.2929803431034088,
1012
+ "selection_signal": "symbol:putdown",
1013
+ "also_edited_in_v1": false,
1014
+ "changed_scalars": 1024,
1015
+ "relative_l2_change": 0.00983031652867794
1016
+ },
1017
+ {
1018
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1019
+ "layer": 16,
1020
+ "column": 3410,
1021
+ "static_cosine": 0.18069753050804138,
1022
+ "selection_signal": "symbol:stack",
1023
+ "also_edited_in_v1": false,
1024
+ "changed_scalars": 1024,
1025
+ "relative_l2_change": 0.010171130299568176
1026
+ },
1027
+ {
1028
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1029
+ "layer": 21,
1030
+ "column": 3350,
1031
+ "static_cosine": 0.31719616055488586,
1032
+ "selection_signal": "symbol:unstack",
1033
+ "also_edited_in_v1": false,
1034
+ "changed_scalars": 1024,
1035
+ "relative_l2_change": 0.009989721700549126
1036
+ },
1037
+ {
1038
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1039
+ "layer": 21,
1040
+ "column": 844,
1041
+ "static_cosine": 0.19255313277244568,
1042
+ "selection_signal": "preconditions:pickup",
1043
+ "also_edited_in_v1": false,
1044
+ "changed_scalars": 1024,
1045
+ "relative_l2_change": 0.009900392033159733
1046
+ },
1047
+ {
1048
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
1049
+ "layer": 20,
1050
+ "column": 195,
1051
+ "static_cosine": 0.18775776028633118,
1052
+ "selection_signal": "effects:pickup",
1053
+ "also_edited_in_v1": false,
1054
+ "changed_scalars": 1024,
1055
+ "relative_l2_change": 0.009959930554032326
1056
+ },
1057
+ {
1058
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1059
+ "layer": 22,
1060
+ "column": 923,
1061
+ "static_cosine": 0.17150506377220154,
1062
+ "selection_signal": "preconditions:putdown",
1063
+ "also_edited_in_v1": false,
1064
+ "changed_scalars": 1024,
1065
+ "relative_l2_change": 0.009932009503245354
1066
+ },
1067
+ {
1068
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1069
+ "layer": 19,
1070
+ "column": 2190,
1071
+ "static_cosine": 0.20565304160118103,
1072
+ "selection_signal": "effects:putdown",
1073
+ "also_edited_in_v1": false,
1074
+ "changed_scalars": 1024,
1075
+ "relative_l2_change": 0.009916073642671108
1076
+ },
1077
+ {
1078
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
1079
+ "layer": 17,
1080
+ "column": 2621,
1081
+ "static_cosine": 0.22644296288490295,
1082
+ "selection_signal": "preconditions:stack",
1083
+ "also_edited_in_v1": false,
1084
+ "changed_scalars": 1024,
1085
+ "relative_l2_change": 0.010032542049884796
1086
+ },
1087
+ {
1088
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1089
+ "layer": 22,
1090
+ "column": 3387,
1091
+ "static_cosine": 0.19442510604858398,
1092
+ "selection_signal": "effects:stack",
1093
+ "also_edited_in_v1": false,
1094
+ "changed_scalars": 1024,
1095
+ "relative_l2_change": 0.01000154484063387
1096
+ },
1097
+ {
1098
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
1099
+ "layer": 18,
1100
+ "column": 1896,
1101
+ "static_cosine": 0.1281166672706604,
1102
+ "selection_signal": "preconditions:unstack",
1103
+ "also_edited_in_v1": false,
1104
+ "changed_scalars": 1024,
1105
+ "relative_l2_change": 0.009758836589753628
1106
+ },
1107
+ {
1108
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1109
+ "layer": 16,
1110
+ "column": 1491,
1111
+ "static_cosine": 0.18250513076782227,
1112
+ "selection_signal": "effects:unstack",
1113
+ "also_edited_in_v1": false,
1114
+ "changed_scalars": 1024,
1115
+ "relative_l2_change": 0.009859840385615826
1116
+ },
1117
+ {
1118
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
1119
+ "layer": 17,
1120
+ "column": 1229,
1121
+ "static_cosine": 0.14222221076488495,
1122
+ "selection_signal": "symbol:clear",
1123
+ "also_edited_in_v1": false,
1124
+ "changed_scalars": 1024,
1125
+ "relative_l2_change": 0.010186979547142982
1126
+ },
1127
+ {
1128
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
1129
+ "layer": 18,
1130
+ "column": 2031,
1131
+ "static_cosine": 0.19438748061656952,
1132
+ "selection_signal": "symbol:on-table",
1133
+ "also_edited_in_v1": false,
1134
+ "changed_scalars": 1024,
1135
+ "relative_l2_change": 0.009989450685679913
1136
+ },
1137
+ {
1138
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
1139
+ "layer": 17,
1140
+ "column": 2072,
1141
+ "static_cosine": 0.14736895263195038,
1142
+ "selection_signal": "symbol:arm-empty",
1143
+ "also_edited_in_v1": false,
1144
+ "changed_scalars": 1024,
1145
+ "relative_l2_change": 0.009858185425400734
1146
+ },
1147
+ {
1148
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
1149
+ "layer": 18,
1150
+ "column": 2215,
1151
+ "static_cosine": 0.18221721053123474,
1152
+ "selection_signal": "symbol:holding",
1153
+ "also_edited_in_v1": false,
1154
+ "changed_scalars": 1024,
1155
+ "relative_l2_change": 0.009958900511264801
1156
+ },
1157
+ {
1158
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
1159
+ "layer": 17,
1160
+ "column": 398,
1161
+ "static_cosine": 0.32019585371017456,
1162
+ "selection_signal": "symbol:on",
1163
+ "also_edited_in_v1": false,
1164
+ "changed_scalars": 1024,
1165
+ "relative_l2_change": 0.009971560910344124
1166
+ },
1167
+ {
1168
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
1169
+ "layer": 15,
1170
+ "column": 360,
1171
+ "static_cosine": 0.13981321454048157,
1172
+ "selection_signal": "symbol:pickup",
1173
+ "also_edited_in_v1": false,
1174
+ "changed_scalars": 1024,
1175
+ "relative_l2_change": 0.010002671740949154
1176
+ },
1177
+ {
1178
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
1179
+ "layer": 17,
1180
+ "column": 295,
1181
+ "static_cosine": 0.16777807474136353,
1182
+ "selection_signal": "symbol:putdown",
1183
+ "also_edited_in_v1": false,
1184
+ "changed_scalars": 1024,
1185
+ "relative_l2_change": 0.009863881394267082
1186
+ },
1187
+ {
1188
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
1189
+ "layer": 17,
1190
+ "column": 2716,
1191
+ "static_cosine": 0.179281547665596,
1192
+ "selection_signal": "symbol:stack",
1193
+ "also_edited_in_v1": false,
1194
+ "changed_scalars": 1024,
1195
+ "relative_l2_change": 0.010081829503178596
1196
+ },
1197
+ {
1198
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1199
+ "layer": 16,
1200
+ "column": 2881,
1201
+ "static_cosine": 0.13326811790466309,
1202
+ "selection_signal": "symbol:unstack",
1203
+ "also_edited_in_v1": false,
1204
+ "changed_scalars": 1024,
1205
+ "relative_l2_change": 0.010200344026088715
1206
+ },
1207
+ {
1208
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1209
+ "layer": 16,
1210
+ "column": 2487,
1211
+ "static_cosine": 0.11868029832839966,
1212
+ "selection_signal": "preconditions:pickup",
1213
+ "also_edited_in_v1": false,
1214
+ "changed_scalars": 1024,
1215
+ "relative_l2_change": 0.009920013137161732
1216
+ },
1217
+ {
1218
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
1219
+ "layer": 12,
1220
+ "column": 3459,
1221
+ "static_cosine": 0.11431478708982468,
1222
+ "selection_signal": "effects:pickup",
1223
+ "also_edited_in_v1": false,
1224
+ "changed_scalars": 1024,
1225
+ "relative_l2_change": 0.009912427514791489
1226
+ },
1227
+ {
1228
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
1229
+ "layer": 12,
1230
+ "column": 2795,
1231
+ "static_cosine": 0.11263640969991684,
1232
+ "selection_signal": "preconditions:putdown",
1233
+ "also_edited_in_v1": false,
1234
+ "changed_scalars": 1024,
1235
+ "relative_l2_change": 0.010020975954830647
1236
+ },
1237
+ {
1238
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1239
+ "layer": 16,
1240
+ "column": 2186,
1241
+ "static_cosine": 0.12058049440383911,
1242
+ "selection_signal": "effects:putdown",
1243
+ "also_edited_in_v1": false,
1244
+ "changed_scalars": 1024,
1245
+ "relative_l2_change": 0.010172060690820217
1246
+ },
1247
+ {
1248
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
1249
+ "layer": 14,
1250
+ "column": 1157,
1251
+ "static_cosine": 0.1378193348646164,
1252
+ "selection_signal": "preconditions:stack",
1253
+ "also_edited_in_v1": false,
1254
+ "changed_scalars": 1024,
1255
+ "relative_l2_change": 0.00987866148352623
1256
+ },
1257
+ {
1258
+ "tensor": "model.language_model.layers.13.mlp.down_proj.weight",
1259
+ "layer": 13,
1260
+ "column": 2328,
1261
+ "static_cosine": 0.127530038356781,
1262
+ "selection_signal": "effects:stack",
1263
+ "also_edited_in_v1": false,
1264
+ "changed_scalars": 1024,
1265
+ "relative_l2_change": 0.009956334717571735
1266
+ },
1267
+ {
1268
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1269
+ "layer": 16,
1270
+ "column": 2688,
1271
+ "static_cosine": 0.12455776333808899,
1272
+ "selection_signal": "preconditions:unstack",
1273
+ "also_edited_in_v1": false,
1274
+ "changed_scalars": 1024,
1275
+ "relative_l2_change": 0.009896550327539444
1276
+ },
1277
+ {
1278
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
1279
+ "layer": 15,
1280
+ "column": 3444,
1281
+ "static_cosine": 0.10998205840587616,
1282
+ "selection_signal": "effects:unstack",
1283
+ "also_edited_in_v1": false,
1284
+ "changed_scalars": 1024,
1285
+ "relative_l2_change": 0.010164337232708931
1286
+ }
1287
+ ],
1288
+ "overlap_with_v1_columns": 0,
1289
+ "cumulative_selected_columns": 132,
1290
+ "limitations": [
1291
+ "Static cosine alignment is not evidence of a causal planning mechanism or accuracy gain.",
1292
+ "Subword pooling is uncontextualized and loses token order; context variants change only tokenization.",
1293
+ "Predicate-level precondition/effect vectors ignore variable binding and argument order.",
1294
+ "The same predicate can be added for one object and deleted for another; this representation cannot express that distinction.",
1295
+ "No action semantics or symbolic solver have been compiled into the network.",
1296
+ "Unknown neuron activation and sign, and subsequent layers, can reverse the intended effect.",
1297
+ "The 1% scale, layer range, and number of columns are hand-chosen and unevaluated.",
1298
+ "The initial v1 heuristic remains in this incremental checkpoint; its benefit is also unknown.",
1299
+ "The example problem is used only for symbolic rule validation, not as a neural training or evaluation example.",
1300
+ "The checkpoint may perform worse on Blocksworld or other tasks."
1301
+ ],
1302
+ "verification": {
1303
+ "all_tensors_checked_against_v1": 488,
1304
+ "changed_tensors": 12,
1305
+ "unchanged_tensors": 476,
1306
+ "changed_scalars": 69632,
1307
+ "unselected_weight_bytes_identical_to_v1": true,
1308
+ "v1_checkpoint_unchanged": true,
1309
+ "config_and_tokenizer_files_identical_to_base": true,
1310
+ "edited_sha256": "71abfea5563dd3550791e032f63928334087481732659718b6e41fc7b04786e5",
1311
+ "model_card_present": false
1312
+ }
1313
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors-00001-of-00001.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71abfea5563dd3550791e032f63928334087481732659718b6e41fc7b04786e5
3
+ size 1746942600
model.safetensors.index.json ADDED
@@ -0,0 +1,495 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 1746882752
4
+ },
5
+ "weight_map": {
6
+ "model.language_model.embed_tokens.weight": "model.safetensors-00001-of-00001.safetensors",
7
+ "model.visual.merger.linear_fc1.weight": "model.safetensors-00001-of-00001.safetensors",
8
+ "model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
9
+ "model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
10
+ "model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
11
+ "model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
12
+ "model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
13
+ "model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
14
+ "model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
15
+ "model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
16
+ "model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
17
+ "model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
18
+ "model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
19
+ "model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
20
+ "model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
21
+ "model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
22
+ "model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
23
+ "model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
24
+ "model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
25
+ "model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model.safetensors-00001-of-00001.safetensors",
26
+ "model.language_model.layers.23.self_attn.q_proj.weight": "model.safetensors-00001-of-00001.safetensors",
27
+ "model.language_model.layers.15.self_attn.q_proj.weight": "model.safetensors-00001-of-00001.safetensors",
28
+ "model.language_model.layers.7.self_attn.q_proj.weight": "model.safetensors-00001-of-00001.safetensors",
29
+ "model.language_model.layers.11.self_attn.q_proj.weight": "model.safetensors-00001-of-00001.safetensors",
30
+ "mtp.layers.0.self_attn.q_proj.weight": "model.safetensors-00001-of-00001.safetensors",
31
+ "model.language_model.layers.19.self_attn.q_proj.weight": "model.safetensors-00001-of-00001.safetensors",
32
+ "model.language_model.layers.3.self_attn.q_proj.weight": "model.safetensors-00001-of-00001.safetensors",
33
+ "model.language_model.layers.22.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
34
+ "model.language_model.layers.22.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
35
+ "model.language_model.layers.22.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
36
+ "model.language_model.layers.23.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
37
+ "model.language_model.layers.23.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
38
+ "model.language_model.layers.23.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
39
+ "model.language_model.layers.3.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
40
+ "model.language_model.layers.3.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
41
+ "model.language_model.layers.3.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
42
+ "model.language_model.layers.14.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
43
+ "model.language_model.layers.14.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
44
+ "model.language_model.layers.14.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
45
+ "model.language_model.layers.15.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
46
+ "model.language_model.layers.15.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
47
+ "model.language_model.layers.15.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
48
+ "model.language_model.layers.6.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
49
+ "model.language_model.layers.6.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
50
+ "model.language_model.layers.6.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
51
+ "model.language_model.layers.7.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
52
+ "model.language_model.layers.7.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
53
+ "model.language_model.layers.7.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
54
+ "model.language_model.layers.10.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
55
+ "model.language_model.layers.11.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
56
+ "model.language_model.layers.11.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
57
+ "model.language_model.layers.11.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
58
+ "model.language_model.layers.9.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
59
+ "model.language_model.layers.9.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
60
+ "model.language_model.layers.9.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
61
+ "mtp.layers.0.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
62
+ "mtp.layers.0.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
63
+ "mtp.layers.0.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
64
+ "model.language_model.layers.17.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
65
+ "model.language_model.layers.17.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
66
+ "model.language_model.layers.17.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
67
+ "model.language_model.layers.18.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
68
+ "model.language_model.layers.18.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
69
+ "model.language_model.layers.18.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
70
+ "model.language_model.layers.12.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
71
+ "model.language_model.layers.12.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
72
+ "model.language_model.layers.12.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
73
+ "model.language_model.layers.13.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
74
+ "model.language_model.layers.13.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
75
+ "model.language_model.layers.13.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
76
+ "model.language_model.layers.19.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
77
+ "model.language_model.layers.19.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
78
+ "model.language_model.layers.19.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
79
+ "model.language_model.layers.4.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
80
+ "model.language_model.layers.4.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
81
+ "model.language_model.layers.4.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
82
+ "model.language_model.layers.2.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
83
+ "model.language_model.layers.2.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
84
+ "model.language_model.layers.2.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
85
+ "model.language_model.layers.20.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
86
+ "model.language_model.layers.20.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
87
+ "model.language_model.layers.20.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
88
+ "model.language_model.layers.1.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
89
+ "model.language_model.layers.1.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
90
+ "model.language_model.layers.1.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
91
+ "model.language_model.layers.10.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
92
+ "model.language_model.layers.10.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
93
+ "model.language_model.layers.16.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
94
+ "model.language_model.layers.16.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
95
+ "model.language_model.layers.16.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
96
+ "model.language_model.layers.5.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
97
+ "model.language_model.layers.5.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
98
+ "model.language_model.layers.5.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
99
+ "model.language_model.layers.8.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
100
+ "model.language_model.layers.8.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
101
+ "model.language_model.layers.8.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
102
+ "model.language_model.layers.21.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
103
+ "model.language_model.layers.21.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
104
+ "model.language_model.layers.21.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
105
+ "model.language_model.layers.0.mlp.down_proj.weight": "model.safetensors-00001-of-00001.safetensors",
106
+ "model.language_model.layers.0.mlp.gate_proj.weight": "model.safetensors-00001-of-00001.safetensors",
107
+ "model.language_model.layers.0.mlp.up_proj.weight": "model.safetensors-00001-of-00001.safetensors",
108
+ "model.visual.merger.linear_fc2.weight": "model.safetensors-00001-of-00001.safetensors",
109
+ "model.visual.blocks.0.mlp.linear_fc1.weight": "model.safetensors-00001-of-00001.safetensors",
110
+ "model.visual.blocks.0.mlp.linear_fc2.weight": "model.safetensors-00001-of-00001.safetensors",
111
+ "model.visual.blocks.1.mlp.linear_fc1.weight": "model.safetensors-00001-of-00001.safetensors",
112
+ "model.visual.blocks.1.mlp.linear_fc2.weight": "model.safetensors-00001-of-00001.safetensors",
113
+ "model.visual.blocks.10.mlp.linear_fc1.weight": "model.safetensors-00001-of-00001.safetensors",
114
+ "model.visual.blocks.10.mlp.linear_fc2.weight": "model.safetensors-00001-of-00001.safetensors",
115
+ "model.visual.blocks.11.mlp.linear_fc1.weight": "model.safetensors-00001-of-00001.safetensors",
116
+ "model.visual.blocks.11.mlp.linear_fc2.weight": "model.safetensors-00001-of-00001.safetensors",
117
+ "model.visual.blocks.2.mlp.linear_fc1.weight": "model.safetensors-00001-of-00001.safetensors",
118
+ "model.visual.blocks.2.mlp.linear_fc2.weight": "model.safetensors-00001-of-00001.safetensors",
119
+ "model.visual.blocks.3.mlp.linear_fc1.weight": "model.safetensors-00001-of-00001.safetensors",
120
+ "model.visual.blocks.3.mlp.linear_fc2.weight": "model.safetensors-00001-of-00001.safetensors",
121
+ "model.visual.blocks.4.mlp.linear_fc1.weight": "model.safetensors-00001-of-00001.safetensors",
122
+ "model.visual.blocks.4.mlp.linear_fc2.weight": "model.safetensors-00001-of-00001.safetensors",
123
+ "model.visual.blocks.5.mlp.linear_fc1.weight": "model.safetensors-00001-of-00001.safetensors",
124
+ "model.visual.blocks.5.mlp.linear_fc2.weight": "model.safetensors-00001-of-00001.safetensors",
125
+ "model.visual.blocks.6.mlp.linear_fc1.weight": "model.safetensors-00001-of-00001.safetensors",
126
+ "model.visual.blocks.6.mlp.linear_fc2.weight": "model.safetensors-00001-of-00001.safetensors",
127
+ "model.visual.blocks.7.mlp.linear_fc1.weight": "model.safetensors-00001-of-00001.safetensors",
128
+ "model.visual.blocks.7.mlp.linear_fc2.weight": "model.safetensors-00001-of-00001.safetensors",
129
+ "model.visual.blocks.8.mlp.linear_fc1.weight": "model.safetensors-00001-of-00001.safetensors",
130
+ "model.visual.blocks.8.mlp.linear_fc2.weight": "model.safetensors-00001-of-00001.safetensors",
131
+ "model.visual.blocks.9.mlp.linear_fc1.weight": "model.safetensors-00001-of-00001.safetensors",
132
+ "model.visual.blocks.9.mlp.linear_fc2.weight": "model.safetensors-00001-of-00001.safetensors",
133
+ "model.language_model.layers.23.self_attn.o_proj.weight": "model.safetensors-00001-of-00001.safetensors",
134
+ "model.language_model.layers.3.self_attn.o_proj.weight": "model.safetensors-00001-of-00001.safetensors",
135
+ "model.language_model.layers.14.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
136
+ "model.language_model.layers.14.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
137
+ "model.language_model.layers.15.self_attn.o_proj.weight": "model.safetensors-00001-of-00001.safetensors",
138
+ "model.language_model.layers.6.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
139
+ "model.language_model.layers.7.self_attn.o_proj.weight": "model.safetensors-00001-of-00001.safetensors",
140
+ "model.language_model.layers.11.self_attn.o_proj.weight": "model.safetensors-00001-of-00001.safetensors",
141
+ "model.language_model.layers.12.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
142
+ "model.language_model.layers.12.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
143
+ "mtp.fc.weight": "model.safetensors-00001-of-00001.safetensors",
144
+ "mtp.layers.0.self_attn.o_proj.weight": "model.safetensors-00001-of-00001.safetensors",
145
+ "model.language_model.layers.17.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
146
+ "model.language_model.layers.18.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
147
+ "model.language_model.layers.18.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
148
+ "model.language_model.layers.13.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
149
+ "model.language_model.layers.13.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
150
+ "model.language_model.layers.19.self_attn.o_proj.weight": "model.safetensors-00001-of-00001.safetensors",
151
+ "model.language_model.layers.2.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
152
+ "model.language_model.layers.2.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
153
+ "model.language_model.layers.4.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
154
+ "model.language_model.layers.4.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
155
+ "model.language_model.layers.20.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
156
+ "model.language_model.layers.20.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
157
+ "model.language_model.layers.1.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
158
+ "model.language_model.layers.1.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
159
+ "model.language_model.layers.10.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
160
+ "model.language_model.layers.10.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
161
+ "model.language_model.layers.16.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
162
+ "model.language_model.layers.16.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
163
+ "model.language_model.layers.17.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
164
+ "model.language_model.layers.5.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
165
+ "model.language_model.layers.5.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
166
+ "model.language_model.layers.6.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
167
+ "model.language_model.layers.8.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
168
+ "model.language_model.layers.8.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
169
+ "model.language_model.layers.9.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
170
+ "model.language_model.layers.9.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
171
+ "model.language_model.layers.21.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
172
+ "model.language_model.layers.21.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
173
+ "model.language_model.layers.22.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
174
+ "model.language_model.layers.22.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
175
+ "model.language_model.layers.0.linear_attn.in_proj_z.weight": "model.safetensors-00001-of-00001.safetensors",
176
+ "model.language_model.layers.0.linear_attn.out_proj.weight": "model.safetensors-00001-of-00001.safetensors",
177
+ "model.visual.blocks.0.attn.qkv.weight": "model.safetensors-00001-of-00001.safetensors",
178
+ "model.visual.blocks.1.attn.qkv.weight": "model.safetensors-00001-of-00001.safetensors",
179
+ "model.visual.blocks.10.attn.qkv.weight": "model.safetensors-00001-of-00001.safetensors",
180
+ "model.visual.blocks.11.attn.qkv.weight": "model.safetensors-00001-of-00001.safetensors",
181
+ "model.visual.blocks.2.attn.qkv.weight": "model.safetensors-00001-of-00001.safetensors",
182
+ "model.visual.blocks.3.attn.qkv.weight": "model.safetensors-00001-of-00001.safetensors",
183
+ "model.visual.blocks.4.attn.qkv.weight": "model.safetensors-00001-of-00001.safetensors",
184
+ "model.visual.blocks.5.attn.qkv.weight": "model.safetensors-00001-of-00001.safetensors",
185
+ "model.visual.blocks.6.attn.qkv.weight": "model.safetensors-00001-of-00001.safetensors",
186
+ "model.visual.blocks.7.attn.qkv.weight": "model.safetensors-00001-of-00001.safetensors",
187
+ "model.visual.blocks.8.attn.qkv.weight": "model.safetensors-00001-of-00001.safetensors",
188
+ "model.visual.blocks.9.attn.qkv.weight": "model.safetensors-00001-of-00001.safetensors",
189
+ "model.visual.pos_embed.weight": "model.safetensors-00001-of-00001.safetensors",
190
+ "model.visual.patch_embed.proj.weight": "model.safetensors-00001-of-00001.safetensors",
191
+ "model.visual.blocks.0.attn.proj.weight": "model.safetensors-00001-of-00001.safetensors",
192
+ "model.visual.blocks.1.attn.proj.weight": "model.safetensors-00001-of-00001.safetensors",
193
+ "model.visual.blocks.10.attn.proj.weight": "model.safetensors-00001-of-00001.safetensors",
194
+ "model.visual.blocks.11.attn.proj.weight": "model.safetensors-00001-of-00001.safetensors",
195
+ "model.visual.blocks.2.attn.proj.weight": "model.safetensors-00001-of-00001.safetensors",
196
+ "model.visual.blocks.3.attn.proj.weight": "model.safetensors-00001-of-00001.safetensors",
197
+ "model.visual.blocks.4.attn.proj.weight": "model.safetensors-00001-of-00001.safetensors",
198
+ "model.visual.blocks.5.attn.proj.weight": "model.safetensors-00001-of-00001.safetensors",
199
+ "model.visual.blocks.6.attn.proj.weight": "model.safetensors-00001-of-00001.safetensors",
200
+ "model.visual.blocks.7.attn.proj.weight": "model.safetensors-00001-of-00001.safetensors",
201
+ "model.visual.blocks.8.attn.proj.weight": "model.safetensors-00001-of-00001.safetensors",
202
+ "model.visual.blocks.9.attn.proj.weight": "model.safetensors-00001-of-00001.safetensors",
203
+ "model.language_model.layers.23.self_attn.k_proj.weight": "model.safetensors-00001-of-00001.safetensors",
204
+ "model.language_model.layers.23.self_attn.v_proj.weight": "model.safetensors-00001-of-00001.safetensors",
205
+ "model.language_model.layers.15.self_attn.k_proj.weight": "model.safetensors-00001-of-00001.safetensors",
206
+ "model.language_model.layers.15.self_attn.v_proj.weight": "model.safetensors-00001-of-00001.safetensors",
207
+ "model.language_model.layers.7.self_attn.k_proj.weight": "model.safetensors-00001-of-00001.safetensors",
208
+ "model.language_model.layers.7.self_attn.v_proj.weight": "model.safetensors-00001-of-00001.safetensors",
209
+ "model.language_model.layers.11.self_attn.k_proj.weight": "model.safetensors-00001-of-00001.safetensors",
210
+ "model.language_model.layers.11.self_attn.v_proj.weight": "model.safetensors-00001-of-00001.safetensors",
211
+ "mtp.layers.0.self_attn.k_proj.weight": "model.safetensors-00001-of-00001.safetensors",
212
+ "mtp.layers.0.self_attn.v_proj.weight": "model.safetensors-00001-of-00001.safetensors",
213
+ "model.language_model.layers.19.self_attn.k_proj.weight": "model.safetensors-00001-of-00001.safetensors",
214
+ "model.language_model.layers.19.self_attn.v_proj.weight": "model.safetensors-00001-of-00001.safetensors",
215
+ "model.language_model.layers.3.self_attn.k_proj.weight": "model.safetensors-00001-of-00001.safetensors",
216
+ "model.language_model.layers.3.self_attn.v_proj.weight": "model.safetensors-00001-of-00001.safetensors",
217
+ "model.language_model.layers.8.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
218
+ "model.language_model.layers.12.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
219
+ "model.language_model.layers.18.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
220
+ "model.language_model.layers.13.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
221
+ "model.language_model.layers.14.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
222
+ "model.language_model.layers.2.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
223
+ "model.language_model.layers.4.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
224
+ "model.language_model.layers.5.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
225
+ "model.language_model.layers.20.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
226
+ "model.language_model.layers.21.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
227
+ "model.language_model.layers.10.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
228
+ "model.language_model.layers.16.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
229
+ "model.language_model.layers.17.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
230
+ "model.language_model.layers.6.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
231
+ "model.language_model.layers.9.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
232
+ "model.language_model.layers.22.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
233
+ "model.language_model.layers.0.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
234
+ "model.language_model.layers.1.linear_attn.conv1d.weight": "model.safetensors-00001-of-00001.safetensors",
235
+ "model.language_model.layers.14.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
236
+ "model.language_model.layers.14.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
237
+ "model.language_model.layers.12.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
238
+ "model.language_model.layers.12.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
239
+ "model.language_model.layers.18.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
240
+ "model.language_model.layers.18.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
241
+ "model.language_model.layers.13.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
242
+ "model.language_model.layers.13.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
243
+ "model.language_model.layers.2.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
244
+ "model.language_model.layers.2.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
245
+ "model.language_model.layers.4.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
246
+ "model.language_model.layers.4.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
247
+ "model.language_model.layers.20.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
248
+ "model.language_model.layers.20.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
249
+ "model.language_model.layers.1.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
250
+ "model.language_model.layers.1.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
251
+ "model.language_model.layers.10.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
252
+ "model.language_model.layers.10.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
253
+ "model.language_model.layers.16.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
254
+ "model.language_model.layers.16.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
255
+ "model.language_model.layers.17.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
256
+ "model.language_model.layers.17.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
257
+ "model.language_model.layers.5.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
258
+ "model.language_model.layers.5.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
259
+ "model.language_model.layers.6.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
260
+ "model.language_model.layers.6.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
261
+ "model.language_model.layers.8.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
262
+ "model.language_model.layers.8.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
263
+ "model.language_model.layers.9.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
264
+ "model.language_model.layers.9.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
265
+ "model.language_model.layers.21.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
266
+ "model.language_model.layers.21.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
267
+ "model.language_model.layers.22.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
268
+ "model.language_model.layers.22.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
269
+ "model.language_model.layers.0.linear_attn.in_proj_b.weight": "model.safetensors-00001-of-00001.safetensors",
270
+ "model.language_model.layers.0.linear_attn.in_proj_a.weight": "model.safetensors-00001-of-00001.safetensors",
271
+ "model.visual.blocks.0.mlp.linear_fc1.bias": "model.safetensors-00001-of-00001.safetensors",
272
+ "model.visual.blocks.1.mlp.linear_fc1.bias": "model.safetensors-00001-of-00001.safetensors",
273
+ "model.visual.blocks.10.mlp.linear_fc1.bias": "model.safetensors-00001-of-00001.safetensors",
274
+ "model.visual.blocks.11.mlp.linear_fc1.bias": "model.safetensors-00001-of-00001.safetensors",
275
+ "model.visual.blocks.2.mlp.linear_fc1.bias": "model.safetensors-00001-of-00001.safetensors",
276
+ "model.visual.blocks.3.mlp.linear_fc1.bias": "model.safetensors-00001-of-00001.safetensors",
277
+ "model.visual.blocks.4.mlp.linear_fc1.bias": "model.safetensors-00001-of-00001.safetensors",
278
+ "model.visual.blocks.5.mlp.linear_fc1.bias": "model.safetensors-00001-of-00001.safetensors",
279
+ "model.visual.blocks.6.mlp.linear_fc1.bias": "model.safetensors-00001-of-00001.safetensors",
280
+ "model.visual.blocks.7.mlp.linear_fc1.bias": "model.safetensors-00001-of-00001.safetensors",
281
+ "model.visual.blocks.8.mlp.linear_fc1.bias": "model.safetensors-00001-of-00001.safetensors",
282
+ "model.visual.blocks.9.mlp.linear_fc1.bias": "model.safetensors-00001-of-00001.safetensors",
283
+ "model.visual.merger.linear_fc1.bias": "model.safetensors-00001-of-00001.safetensors",
284
+ "model.visual.blocks.0.attn.qkv.bias": "model.safetensors-00001-of-00001.safetensors",
285
+ "model.visual.blocks.1.attn.qkv.bias": "model.safetensors-00001-of-00001.safetensors",
286
+ "model.visual.blocks.10.attn.qkv.bias": "model.safetensors-00001-of-00001.safetensors",
287
+ "model.visual.blocks.11.attn.qkv.bias": "model.safetensors-00001-of-00001.safetensors",
288
+ "model.visual.blocks.2.attn.qkv.bias": "model.safetensors-00001-of-00001.safetensors",
289
+ "model.visual.blocks.3.attn.qkv.bias": "model.safetensors-00001-of-00001.safetensors",
290
+ "model.visual.blocks.4.attn.qkv.bias": "model.safetensors-00001-of-00001.safetensors",
291
+ "model.visual.blocks.5.attn.qkv.bias": "model.safetensors-00001-of-00001.safetensors",
292
+ "model.visual.blocks.6.attn.qkv.bias": "model.safetensors-00001-of-00001.safetensors",
293
+ "model.visual.blocks.7.attn.qkv.bias": "model.safetensors-00001-of-00001.safetensors",
294
+ "model.visual.blocks.8.attn.qkv.bias": "model.safetensors-00001-of-00001.safetensors",
295
+ "model.visual.blocks.9.attn.qkv.bias": "model.safetensors-00001-of-00001.safetensors",
296
+ "model.language_model.layers.22.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
297
+ "model.language_model.layers.23.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
298
+ "model.language_model.layers.23.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
299
+ "model.language_model.layers.3.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
300
+ "model.language_model.layers.14.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
301
+ "model.language_model.layers.14.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
302
+ "model.language_model.layers.15.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
303
+ "model.language_model.layers.15.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
304
+ "model.language_model.layers.6.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
305
+ "model.language_model.layers.7.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
306
+ "model.language_model.layers.7.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
307
+ "model.language_model.layers.8.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
308
+ "model.language_model.layers.11.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
309
+ "model.language_model.layers.11.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
310
+ "model.language_model.layers.12.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
311
+ "mtp.layers.0.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
312
+ "mtp.layers.0.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
313
+ "mtp.norm.weight": "model.safetensors-00001-of-00001.safetensors",
314
+ "mtp.pre_fc_norm_embedding.weight": "model.safetensors-00001-of-00001.safetensors",
315
+ "mtp.pre_fc_norm_hidden.weight": "model.safetensors-00001-of-00001.safetensors",
316
+ "model.language_model.layers.17.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
317
+ "model.language_model.layers.18.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
318
+ "model.language_model.layers.18.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
319
+ "model.language_model.layers.19.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
320
+ "model.language_model.layers.12.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
321
+ "model.language_model.layers.13.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
322
+ "model.language_model.layers.13.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
323
+ "model.language_model.layers.19.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
324
+ "model.language_model.layers.2.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
325
+ "model.language_model.layers.2.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
326
+ "model.language_model.layers.3.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
327
+ "model.language_model.layers.4.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
328
+ "model.language_model.layers.4.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
329
+ "model.language_model.layers.20.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
330
+ "model.language_model.layers.20.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
331
+ "model.language_model.layers.1.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
332
+ "model.language_model.layers.10.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
333
+ "model.language_model.layers.10.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
334
+ "model.language_model.layers.16.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
335
+ "model.language_model.layers.16.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
336
+ "model.language_model.layers.17.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
337
+ "model.language_model.layers.5.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
338
+ "model.language_model.layers.5.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
339
+ "model.language_model.layers.6.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
340
+ "model.language_model.layers.8.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
341
+ "model.language_model.layers.9.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
342
+ "model.language_model.layers.9.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
343
+ "model.language_model.layers.21.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
344
+ "model.language_model.layers.21.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
345
+ "model.language_model.layers.22.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
346
+ "model.language_model.layers.0.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
347
+ "model.language_model.layers.0.post_attention_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
348
+ "model.language_model.layers.1.input_layernorm.weight": "model.safetensors-00001-of-00001.safetensors",
349
+ "model.language_model.norm.weight": "model.safetensors-00001-of-00001.safetensors",
350
+ "model.visual.merger.linear_fc2.bias": "model.safetensors-00001-of-00001.safetensors",
351
+ "model.visual.blocks.0.attn.proj.bias": "model.safetensors-00001-of-00001.safetensors",
352
+ "model.visual.blocks.0.mlp.linear_fc2.bias": "model.safetensors-00001-of-00001.safetensors",
353
+ "model.visual.blocks.0.norm1.bias": "model.safetensors-00001-of-00001.safetensors",
354
+ "model.visual.blocks.0.norm1.weight": "model.safetensors-00001-of-00001.safetensors",
355
+ "model.visual.blocks.0.norm2.bias": "model.safetensors-00001-of-00001.safetensors",
356
+ "model.visual.blocks.0.norm2.weight": "model.safetensors-00001-of-00001.safetensors",
357
+ "model.visual.blocks.1.attn.proj.bias": "model.safetensors-00001-of-00001.safetensors",
358
+ "model.visual.blocks.1.mlp.linear_fc2.bias": "model.safetensors-00001-of-00001.safetensors",
359
+ "model.visual.blocks.1.norm1.bias": "model.safetensors-00001-of-00001.safetensors",
360
+ "model.visual.blocks.1.norm1.weight": "model.safetensors-00001-of-00001.safetensors",
361
+ "model.visual.blocks.1.norm2.bias": "model.safetensors-00001-of-00001.safetensors",
362
+ "model.visual.blocks.1.norm2.weight": "model.safetensors-00001-of-00001.safetensors",
363
+ "model.visual.blocks.10.attn.proj.bias": "model.safetensors-00001-of-00001.safetensors",
364
+ "model.visual.blocks.10.mlp.linear_fc2.bias": "model.safetensors-00001-of-00001.safetensors",
365
+ "model.visual.blocks.10.norm1.bias": "model.safetensors-00001-of-00001.safetensors",
366
+ "model.visual.blocks.10.norm1.weight": "model.safetensors-00001-of-00001.safetensors",
367
+ "model.visual.blocks.10.norm2.bias": "model.safetensors-00001-of-00001.safetensors",
368
+ "model.visual.blocks.10.norm2.weight": "model.safetensors-00001-of-00001.safetensors",
369
+ "model.visual.blocks.11.attn.proj.bias": "model.safetensors-00001-of-00001.safetensors",
370
+ "model.visual.blocks.11.mlp.linear_fc2.bias": "model.safetensors-00001-of-00001.safetensors",
371
+ "model.visual.blocks.11.norm1.bias": "model.safetensors-00001-of-00001.safetensors",
372
+ "model.visual.blocks.11.norm1.weight": "model.safetensors-00001-of-00001.safetensors",
373
+ "model.visual.blocks.11.norm2.bias": "model.safetensors-00001-of-00001.safetensors",
374
+ "model.visual.blocks.11.norm2.weight": "model.safetensors-00001-of-00001.safetensors",
375
+ "model.visual.blocks.2.attn.proj.bias": "model.safetensors-00001-of-00001.safetensors",
376
+ "model.visual.blocks.2.mlp.linear_fc2.bias": "model.safetensors-00001-of-00001.safetensors",
377
+ "model.visual.blocks.2.norm1.bias": "model.safetensors-00001-of-00001.safetensors",
378
+ "model.visual.blocks.2.norm1.weight": "model.safetensors-00001-of-00001.safetensors",
379
+ "model.visual.blocks.2.norm2.bias": "model.safetensors-00001-of-00001.safetensors",
380
+ "model.visual.blocks.2.norm2.weight": "model.safetensors-00001-of-00001.safetensors",
381
+ "model.visual.blocks.3.attn.proj.bias": "model.safetensors-00001-of-00001.safetensors",
382
+ "model.visual.blocks.3.mlp.linear_fc2.bias": "model.safetensors-00001-of-00001.safetensors",
383
+ "model.visual.blocks.3.norm1.bias": "model.safetensors-00001-of-00001.safetensors",
384
+ "model.visual.blocks.3.norm1.weight": "model.safetensors-00001-of-00001.safetensors",
385
+ "model.visual.blocks.3.norm2.bias": "model.safetensors-00001-of-00001.safetensors",
386
+ "model.visual.blocks.3.norm2.weight": "model.safetensors-00001-of-00001.safetensors",
387
+ "model.visual.blocks.4.attn.proj.bias": "model.safetensors-00001-of-00001.safetensors",
388
+ "model.visual.blocks.4.mlp.linear_fc2.bias": "model.safetensors-00001-of-00001.safetensors",
389
+ "model.visual.blocks.4.norm1.bias": "model.safetensors-00001-of-00001.safetensors",
390
+ "model.visual.blocks.4.norm1.weight": "model.safetensors-00001-of-00001.safetensors",
391
+ "model.visual.blocks.4.norm2.bias": "model.safetensors-00001-of-00001.safetensors",
392
+ "model.visual.blocks.4.norm2.weight": "model.safetensors-00001-of-00001.safetensors",
393
+ "model.visual.blocks.5.attn.proj.bias": "model.safetensors-00001-of-00001.safetensors",
394
+ "model.visual.blocks.5.mlp.linear_fc2.bias": "model.safetensors-00001-of-00001.safetensors",
395
+ "model.visual.blocks.5.norm1.bias": "model.safetensors-00001-of-00001.safetensors",
396
+ "model.visual.blocks.5.norm1.weight": "model.safetensors-00001-of-00001.safetensors",
397
+ "model.visual.blocks.5.norm2.bias": "model.safetensors-00001-of-00001.safetensors",
398
+ "model.visual.blocks.5.norm2.weight": "model.safetensors-00001-of-00001.safetensors",
399
+ "model.visual.blocks.6.attn.proj.bias": "model.safetensors-00001-of-00001.safetensors",
400
+ "model.visual.blocks.6.mlp.linear_fc2.bias": "model.safetensors-00001-of-00001.safetensors",
401
+ "model.visual.blocks.6.norm1.bias": "model.safetensors-00001-of-00001.safetensors",
402
+ "model.visual.blocks.6.norm1.weight": "model.safetensors-00001-of-00001.safetensors",
403
+ "model.visual.blocks.6.norm2.bias": "model.safetensors-00001-of-00001.safetensors",
404
+ "model.visual.blocks.6.norm2.weight": "model.safetensors-00001-of-00001.safetensors",
405
+ "model.visual.blocks.7.attn.proj.bias": "model.safetensors-00001-of-00001.safetensors",
406
+ "model.visual.blocks.7.mlp.linear_fc2.bias": "model.safetensors-00001-of-00001.safetensors",
407
+ "model.visual.blocks.7.norm1.bias": "model.safetensors-00001-of-00001.safetensors",
408
+ "model.visual.blocks.7.norm1.weight": "model.safetensors-00001-of-00001.safetensors",
409
+ "model.visual.blocks.7.norm2.bias": "model.safetensors-00001-of-00001.safetensors",
410
+ "model.visual.blocks.7.norm2.weight": "model.safetensors-00001-of-00001.safetensors",
411
+ "model.visual.blocks.8.attn.proj.bias": "model.safetensors-00001-of-00001.safetensors",
412
+ "model.visual.blocks.8.mlp.linear_fc2.bias": "model.safetensors-00001-of-00001.safetensors",
413
+ "model.visual.blocks.8.norm1.bias": "model.safetensors-00001-of-00001.safetensors",
414
+ "model.visual.blocks.8.norm1.weight": "model.safetensors-00001-of-00001.safetensors",
415
+ "model.visual.blocks.8.norm2.bias": "model.safetensors-00001-of-00001.safetensors",
416
+ "model.visual.blocks.8.norm2.weight": "model.safetensors-00001-of-00001.safetensors",
417
+ "model.visual.blocks.9.attn.proj.bias": "model.safetensors-00001-of-00001.safetensors",
418
+ "model.visual.blocks.9.mlp.linear_fc2.bias": "model.safetensors-00001-of-00001.safetensors",
419
+ "model.visual.blocks.9.norm1.bias": "model.safetensors-00001-of-00001.safetensors",
420
+ "model.visual.blocks.9.norm1.weight": "model.safetensors-00001-of-00001.safetensors",
421
+ "model.visual.blocks.9.norm2.bias": "model.safetensors-00001-of-00001.safetensors",
422
+ "model.visual.blocks.9.norm2.weight": "model.safetensors-00001-of-00001.safetensors",
423
+ "model.visual.merger.norm.bias": "model.safetensors-00001-of-00001.safetensors",
424
+ "model.visual.merger.norm.weight": "model.safetensors-00001-of-00001.safetensors",
425
+ "model.visual.patch_embed.proj.bias": "model.safetensors-00001-of-00001.safetensors",
426
+ "model.language_model.layers.23.self_attn.k_norm.weight": "model.safetensors-00001-of-00001.safetensors",
427
+ "model.language_model.layers.23.self_attn.q_norm.weight": "model.safetensors-00001-of-00001.safetensors",
428
+ "model.language_model.layers.3.self_attn.k_norm.weight": "model.safetensors-00001-of-00001.safetensors",
429
+ "model.language_model.layers.14.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
430
+ "model.language_model.layers.15.self_attn.k_norm.weight": "model.safetensors-00001-of-00001.safetensors",
431
+ "model.language_model.layers.15.self_attn.q_norm.weight": "model.safetensors-00001-of-00001.safetensors",
432
+ "model.language_model.layers.7.self_attn.k_norm.weight": "model.safetensors-00001-of-00001.safetensors",
433
+ "model.language_model.layers.7.self_attn.q_norm.weight": "model.safetensors-00001-of-00001.safetensors",
434
+ "model.language_model.layers.11.self_attn.k_norm.weight": "model.safetensors-00001-of-00001.safetensors",
435
+ "model.language_model.layers.11.self_attn.q_norm.weight": "model.safetensors-00001-of-00001.safetensors",
436
+ "model.language_model.layers.12.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
437
+ "mtp.layers.0.self_attn.k_norm.weight": "model.safetensors-00001-of-00001.safetensors",
438
+ "mtp.layers.0.self_attn.q_norm.weight": "model.safetensors-00001-of-00001.safetensors",
439
+ "model.language_model.layers.18.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
440
+ "model.language_model.layers.13.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
441
+ "model.language_model.layers.19.self_attn.k_norm.weight": "model.safetensors-00001-of-00001.safetensors",
442
+ "model.language_model.layers.19.self_attn.q_norm.weight": "model.safetensors-00001-of-00001.safetensors",
443
+ "model.language_model.layers.2.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
444
+ "model.language_model.layers.3.self_attn.q_norm.weight": "model.safetensors-00001-of-00001.safetensors",
445
+ "model.language_model.layers.4.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
446
+ "model.language_model.layers.20.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
447
+ "model.language_model.layers.1.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
448
+ "model.language_model.layers.10.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
449
+ "model.language_model.layers.16.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
450
+ "model.language_model.layers.17.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
451
+ "model.language_model.layers.5.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
452
+ "model.language_model.layers.6.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
453
+ "model.language_model.layers.8.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
454
+ "model.language_model.layers.9.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
455
+ "model.language_model.layers.21.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
456
+ "model.language_model.layers.22.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
457
+ "model.language_model.layers.0.linear_attn.norm.weight": "model.safetensors-00001-of-00001.safetensors",
458
+ "model.language_model.layers.16.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
459
+ "model.language_model.layers.8.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
460
+ "model.language_model.layers.12.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
461
+ "model.language_model.layers.18.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
462
+ "model.language_model.layers.13.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
463
+ "model.language_model.layers.14.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
464
+ "model.language_model.layers.2.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
465
+ "model.language_model.layers.4.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
466
+ "model.language_model.layers.5.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
467
+ "model.language_model.layers.20.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
468
+ "model.language_model.layers.21.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
469
+ "model.language_model.layers.10.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
470
+ "model.language_model.layers.17.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
471
+ "model.language_model.layers.6.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
472
+ "model.language_model.layers.9.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
473
+ "model.language_model.layers.22.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
474
+ "model.language_model.layers.0.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
475
+ "model.language_model.layers.1.linear_attn.A_log": "model.safetensors-00001-of-00001.safetensors",
476
+ "model.language_model.layers.8.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
477
+ "model.language_model.layers.12.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
478
+ "model.language_model.layers.18.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
479
+ "model.language_model.layers.13.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
480
+ "model.language_model.layers.14.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
481
+ "model.language_model.layers.2.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
482
+ "model.language_model.layers.4.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
483
+ "model.language_model.layers.20.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
484
+ "model.language_model.layers.21.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
485
+ "model.language_model.layers.10.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
486
+ "model.language_model.layers.16.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
487
+ "model.language_model.layers.17.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
488
+ "model.language_model.layers.5.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
489
+ "model.language_model.layers.6.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
490
+ "model.language_model.layers.9.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
491
+ "model.language_model.layers.22.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
492
+ "model.language_model.layers.0.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors",
493
+ "model.language_model.layers.1.linear_attn.dt_bias": "model.safetensors-00001-of-00001.safetensors"
494
+ }
495
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "size": {
3
+ "longest_edge": 16777216,
4
+ "shortest_edge": 65536
5
+ },
6
+ "patch_size": 16,
7
+ "temporal_patch_size": 2,
8
+ "merge_size": 2,
9
+ "image_mean": [
10
+ 0.5,
11
+ 0.5,
12
+ 0.5
13
+ ],
14
+ "image_std": [
15
+ 0.5,
16
+ 0.5,
17
+ 0.5
18
+ ],
19
+ "processor_class": "Qwen3VLProcessor",
20
+ "image_processor_type": "Qwen2VLImageProcessorFast"
21
+ }
previous_columns.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:604a9edd93538f40727900c23715ad249f5b4990f11a42f0ba32c3a13228da5d
3
+ size 157746
schema_analysis.json ADDED
@@ -0,0 +1,335 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "predicates": [
3
+ [
4
+ "clear",
5
+ "?x"
6
+ ],
7
+ [
8
+ "on-table",
9
+ "?x"
10
+ ],
11
+ [
12
+ "arm-empty"
13
+ ],
14
+ [
15
+ "holding",
16
+ "?x"
17
+ ],
18
+ [
19
+ "on",
20
+ "?x",
21
+ "?y"
22
+ ]
23
+ ],
24
+ "actions": {
25
+ "pickup": {
26
+ "parameters": [
27
+ "?ob"
28
+ ],
29
+ "preconditions": [
30
+ [
31
+ "clear",
32
+ "?ob"
33
+ ],
34
+ [
35
+ "on-table",
36
+ "?ob"
37
+ ],
38
+ [
39
+ "arm-empty"
40
+ ]
41
+ ],
42
+ "add": [
43
+ [
44
+ "holding",
45
+ "?ob"
46
+ ]
47
+ ],
48
+ "delete": [
49
+ [
50
+ "clear",
51
+ "?ob"
52
+ ],
53
+ [
54
+ "on-table",
55
+ "?ob"
56
+ ],
57
+ [
58
+ "arm-empty"
59
+ ]
60
+ ]
61
+ },
62
+ "putdown": {
63
+ "parameters": [
64
+ "?ob"
65
+ ],
66
+ "preconditions": [
67
+ [
68
+ "holding",
69
+ "?ob"
70
+ ]
71
+ ],
72
+ "add": [
73
+ [
74
+ "clear",
75
+ "?ob"
76
+ ],
77
+ [
78
+ "arm-empty"
79
+ ],
80
+ [
81
+ "on-table",
82
+ "?ob"
83
+ ]
84
+ ],
85
+ "delete": [
86
+ [
87
+ "holding",
88
+ "?ob"
89
+ ]
90
+ ]
91
+ },
92
+ "stack": {
93
+ "parameters": [
94
+ "?ob",
95
+ "?underob"
96
+ ],
97
+ "preconditions": [
98
+ [
99
+ "clear",
100
+ "?underob"
101
+ ],
102
+ [
103
+ "holding",
104
+ "?ob"
105
+ ]
106
+ ],
107
+ "add": [
108
+ [
109
+ "arm-empty"
110
+ ],
111
+ [
112
+ "clear",
113
+ "?ob"
114
+ ],
115
+ [
116
+ "on",
117
+ "?ob",
118
+ "?underob"
119
+ ]
120
+ ],
121
+ "delete": [
122
+ [
123
+ "clear",
124
+ "?underob"
125
+ ],
126
+ [
127
+ "holding",
128
+ "?ob"
129
+ ]
130
+ ]
131
+ },
132
+ "unstack": {
133
+ "parameters": [
134
+ "?ob",
135
+ "?underob"
136
+ ],
137
+ "preconditions": [
138
+ [
139
+ "on",
140
+ "?ob",
141
+ "?underob"
142
+ ],
143
+ [
144
+ "clear",
145
+ "?ob"
146
+ ],
147
+ [
148
+ "arm-empty"
149
+ ]
150
+ ],
151
+ "add": [
152
+ [
153
+ "holding",
154
+ "?ob"
155
+ ],
156
+ [
157
+ "clear",
158
+ "?underob"
159
+ ]
160
+ ],
161
+ "delete": [
162
+ [
163
+ "on",
164
+ "?ob",
165
+ "?underob"
166
+ ],
167
+ [
168
+ "clear",
169
+ "?ob"
170
+ ],
171
+ [
172
+ "arm-empty"
173
+ ]
174
+ ]
175
+ }
176
+ },
177
+ "rules": {
178
+ "reachable_three_block_states": 22,
179
+ "legal_transitions_checked": 42,
180
+ "state_invariants_passed": true,
181
+ "inverse_actions_passed": true
182
+ },
183
+ "example": {
184
+ "goal_satisfied_symbolically": true,
185
+ "plan_length": 32,
186
+ "plan": [
187
+ [
188
+ "unstack",
189
+ "b2",
190
+ "b10"
191
+ ],
192
+ [
193
+ "putdown",
194
+ "b2"
195
+ ],
196
+ [
197
+ "unstack",
198
+ "b10",
199
+ "b4"
200
+ ],
201
+ [
202
+ "putdown",
203
+ "b10"
204
+ ],
205
+ [
206
+ "unstack",
207
+ "b4",
208
+ "b8"
209
+ ],
210
+ [
211
+ "putdown",
212
+ "b4"
213
+ ],
214
+ [
215
+ "unstack",
216
+ "b8",
217
+ "b6"
218
+ ],
219
+ [
220
+ "putdown",
221
+ "b8"
222
+ ],
223
+ [
224
+ "unstack",
225
+ "b6",
226
+ "b3"
227
+ ],
228
+ [
229
+ "putdown",
230
+ "b6"
231
+ ],
232
+ [
233
+ "unstack",
234
+ "b3",
235
+ "b7"
236
+ ],
237
+ [
238
+ "putdown",
239
+ "b3"
240
+ ],
241
+ [
242
+ "unstack",
243
+ "b7",
244
+ "b5"
245
+ ],
246
+ [
247
+ "putdown",
248
+ "b7"
249
+ ],
250
+ [
251
+ "unstack",
252
+ "b5",
253
+ "b1"
254
+ ],
255
+ [
256
+ "putdown",
257
+ "b5"
258
+ ],
259
+ [
260
+ "pickup",
261
+ "b1"
262
+ ],
263
+ [
264
+ "stack",
265
+ "b1",
266
+ "b8"
267
+ ],
268
+ [
269
+ "pickup",
270
+ "b5"
271
+ ],
272
+ [
273
+ "stack",
274
+ "b5",
275
+ "b1"
276
+ ],
277
+ [
278
+ "pickup",
279
+ "b4"
280
+ ],
281
+ [
282
+ "stack",
283
+ "b4",
284
+ "b5"
285
+ ],
286
+ [
287
+ "pickup",
288
+ "b10"
289
+ ],
290
+ [
291
+ "stack",
292
+ "b10",
293
+ "b4"
294
+ ],
295
+ [
296
+ "pickup",
297
+ "b6"
298
+ ],
299
+ [
300
+ "stack",
301
+ "b6",
302
+ "b10"
303
+ ],
304
+ [
305
+ "pickup",
306
+ "b7"
307
+ ],
308
+ [
309
+ "stack",
310
+ "b7",
311
+ "b6"
312
+ ],
313
+ [
314
+ "pickup",
315
+ "b2"
316
+ ],
317
+ [
318
+ "stack",
319
+ "b2",
320
+ "b7"
321
+ ],
322
+ [
323
+ "pickup",
324
+ "b3"
325
+ ],
326
+ [
327
+ "stack",
328
+ "b3",
329
+ "b2"
330
+ ]
331
+ ],
332
+ "model_evaluation": false,
333
+ "optimality_claimed": false
334
+ }
335
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f9e4d4901a92b997e463c1f46055088b6cca5ca61a6522d1b9f64c4bb81cb42
3
+ size 12807982
tokenizer_config.json ADDED
@@ -0,0 +1,305 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "248044": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "248045": {
13
+ "content": "<|im_start|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "248046": {
21
+ "content": "<|im_end|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "248047": {
29
+ "content": "<|object_ref_start|>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "248048": {
37
+ "content": "<|object_ref_end|>",
38
+ "lstrip": false,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ },
44
+ "248049": {
45
+ "content": "<|box_start|>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": true
51
+ },
52
+ "248050": {
53
+ "content": "<|box_end|>",
54
+ "lstrip": false,
55
+ "normalized": false,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": true
59
+ },
60
+ "248051": {
61
+ "content": "<|quad_start|>",
62
+ "lstrip": false,
63
+ "normalized": false,
64
+ "rstrip": false,
65
+ "single_word": false,
66
+ "special": true
67
+ },
68
+ "248052": {
69
+ "content": "<|quad_end|>",
70
+ "lstrip": false,
71
+ "normalized": false,
72
+ "rstrip": false,
73
+ "single_word": false,
74
+ "special": true
75
+ },
76
+ "248053": {
77
+ "content": "<|vision_start|>",
78
+ "lstrip": false,
79
+ "normalized": false,
80
+ "rstrip": false,
81
+ "single_word": false,
82
+ "special": true
83
+ },
84
+ "248054": {
85
+ "content": "<|vision_end|>",
86
+ "lstrip": false,
87
+ "normalized": false,
88
+ "rstrip": false,
89
+ "single_word": false,
90
+ "special": true
91
+ },
92
+ "248055": {
93
+ "content": "<|vision_pad|>",
94
+ "lstrip": false,
95
+ "normalized": false,
96
+ "rstrip": false,
97
+ "single_word": false,
98
+ "special": true
99
+ },
100
+ "248056": {
101
+ "content": "<|image_pad|>",
102
+ "lstrip": false,
103
+ "normalized": false,
104
+ "rstrip": false,
105
+ "single_word": false,
106
+ "special": true
107
+ },
108
+ "248057": {
109
+ "content": "<|video_pad|>",
110
+ "lstrip": false,
111
+ "normalized": false,
112
+ "rstrip": false,
113
+ "single_word": false,
114
+ "special": true
115
+ },
116
+ "248058": {
117
+ "content": "<tool_call>",
118
+ "lstrip": false,
119
+ "normalized": false,
120
+ "rstrip": false,
121
+ "single_word": false,
122
+ "special": false
123
+ },
124
+ "248059": {
125
+ "content": "</tool_call>",
126
+ "lstrip": false,
127
+ "normalized": false,
128
+ "rstrip": false,
129
+ "single_word": false,
130
+ "special": false
131
+ },
132
+ "248060": {
133
+ "content": "<|fim_prefix|>",
134
+ "lstrip": false,
135
+ "normalized": false,
136
+ "rstrip": false,
137
+ "single_word": false,
138
+ "special": false
139
+ },
140
+ "248061": {
141
+ "content": "<|fim_middle|>",
142
+ "lstrip": false,
143
+ "normalized": false,
144
+ "rstrip": false,
145
+ "single_word": false,
146
+ "special": false
147
+ },
148
+ "248062": {
149
+ "content": "<|fim_suffix|>",
150
+ "lstrip": false,
151
+ "normalized": false,
152
+ "rstrip": false,
153
+ "single_word": false,
154
+ "special": false
155
+ },
156
+ "248063": {
157
+ "content": "<|fim_pad|>",
158
+ "lstrip": false,
159
+ "normalized": false,
160
+ "rstrip": false,
161
+ "single_word": false,
162
+ "special": false
163
+ },
164
+ "248064": {
165
+ "content": "<|repo_name|>",
166
+ "lstrip": false,
167
+ "normalized": false,
168
+ "rstrip": false,
169
+ "single_word": false,
170
+ "special": false
171
+ },
172
+ "248065": {
173
+ "content": "<|file_sep|>",
174
+ "lstrip": false,
175
+ "normalized": false,
176
+ "rstrip": false,
177
+ "single_word": false,
178
+ "special": false
179
+ },
180
+ "248066": {
181
+ "content": "<tool_response>",
182
+ "lstrip": false,
183
+ "normalized": false,
184
+ "rstrip": false,
185
+ "single_word": false,
186
+ "special": false
187
+ },
188
+ "248067": {
189
+ "content": "</tool_response>",
190
+ "lstrip": false,
191
+ "normalized": false,
192
+ "rstrip": false,
193
+ "single_word": false,
194
+ "special": false
195
+ },
196
+ "248068": {
197
+ "content": "<think>",
198
+ "lstrip": false,
199
+ "normalized": false,
200
+ "rstrip": false,
201
+ "single_word": false,
202
+ "special": false
203
+ },
204
+ "248069": {
205
+ "content": "</think>",
206
+ "lstrip": false,
207
+ "normalized": false,
208
+ "rstrip": false,
209
+ "single_word": false,
210
+ "special": false
211
+ },
212
+ "248070": {
213
+ "content": "<|audio_start|>",
214
+ "lstrip": false,
215
+ "normalized": false,
216
+ "rstrip": false,
217
+ "single_word": false,
218
+ "special": true
219
+ },
220
+ "248071": {
221
+ "content": "<|audio_end|>",
222
+ "lstrip": false,
223
+ "normalized": false,
224
+ "rstrip": false,
225
+ "single_word": false,
226
+ "special": true
227
+ },
228
+ "248072": {
229
+ "content": "<tts_pad>",
230
+ "lstrip": false,
231
+ "normalized": false,
232
+ "rstrip": false,
233
+ "single_word": false,
234
+ "special": true
235
+ },
236
+ "248073": {
237
+ "content": "<tts_text_bos>",
238
+ "lstrip": false,
239
+ "normalized": false,
240
+ "rstrip": false,
241
+ "single_word": false,
242
+ "special": true
243
+ },
244
+ "248074": {
245
+ "content": "<tts_text_eod>",
246
+ "lstrip": false,
247
+ "normalized": false,
248
+ "rstrip": false,
249
+ "single_word": false,
250
+ "special": true
251
+ },
252
+ "248075": {
253
+ "content": "<tts_text_bos_single>",
254
+ "lstrip": false,
255
+ "normalized": false,
256
+ "rstrip": false,
257
+ "single_word": false,
258
+ "special": true
259
+ },
260
+ "248076": {
261
+ "content": "<|audio_pad|>",
262
+ "lstrip": false,
263
+ "normalized": false,
264
+ "rstrip": false,
265
+ "single_word": false,
266
+ "special": true
267
+ }
268
+ },
269
+ "additional_special_tokens": [
270
+ "<|im_start|>",
271
+ "<|im_end|>",
272
+ "<|object_ref_start|>",
273
+ "<|object_ref_end|>",
274
+ "<|box_start|>",
275
+ "<|box_end|>",
276
+ "<|quad_start|>",
277
+ "<|quad_end|>",
278
+ "<|vision_start|>",
279
+ "<|vision_end|>",
280
+ "<|vision_pad|>",
281
+ "<|image_pad|>",
282
+ "<|video_pad|>"
283
+ ],
284
+ "bos_token": null,
285
+ "chat_template": "{%- set image_count = namespace(value=0) %}\n{%- set video_count = namespace(value=0) %}\n{%- macro render_content(content, do_vision_count, is_system_content=false) %}\n {%- if content is string %}\n {{- content }}\n {%- elif content is iterable and content is not mapping %}\n {%- for item in content %}\n {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}\n {%- if is_system_content %}\n {{- raise_exception('System message cannot contain images.') }}\n {%- endif %}\n {%- if do_vision_count %}\n {%- set image_count.value = image_count.value + 1 %}\n {%- endif %}\n {%- if add_vision_id %}\n {{- 'Picture ' ~ image_count.value ~ ': ' }}\n {%- endif %}\n {{- '<|vision_start|><|image_pad|><|vision_end|>' }}\n {%- elif 'video' in item or item.type == 'video' %}\n {%- if is_system_content %}\n {{- raise_exception('System message cannot contain videos.') }}\n {%- endif %}\n {%- if do_vision_count %}\n {%- set video_count.value = video_count.value + 1 %}\n {%- endif %}\n {%- if add_vision_id %}\n {{- 'Video ' ~ video_count.value ~ ': ' }}\n {%- endif %}\n {{- '<|vision_start|><|video_pad|><|vision_end|>' }}\n {%- elif 'text' in item %}\n {{- item.text }}\n {%- else %}\n {{- raise_exception('Unexpected item type in content.') }}\n {%- endif %}\n {%- endfor %}\n {%- elif content is none or content is undefined %}\n {{- '' }}\n {%- else %}\n {{- raise_exception('Unexpected content type.') }}\n {%- endif %}\n{%- endmacro %}\n{%- if not messages %}\n {{- raise_exception('No messages provided.') }}\n{%- endif %}\n{%- if tools and tools is iterable and tools is not mapping %}\n {{- '<|im_start|>system\\n' }}\n {{- \"# Tools\\n\\nYou have access to the following functions:\\n\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\" }}\n {{- '\\n\\nIf you choose to call a function ONLY reply in the following format with NO suffix:\\n\\n<tool_call>\\n<function=example_function_name>\\n<parameter=example_parameter_1>\\nvalue_1\\n</parameter>\\n<parameter=example_parameter_2>\\nThis is the value for the second parameter\\nthat can span\\nmultiple lines\\n</parameter>\\n</function>\\n</tool_call>\\n\\n<IMPORTANT>\\nReminder:\\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\\n- Required parameters MUST be specified\\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\\n</IMPORTANT>' }}\n {%- if messages[0].role == 'system' %}\n {%- set content = render_content(messages[0].content, false, true)|trim %}\n {%- if content %}\n {{- '\\n\\n' + content }}\n {%- endif %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n{%- else %}\n {%- if messages[0].role == 'system' %}\n {%- set content = render_content(messages[0].content, false, true)|trim %}\n {{- '<|im_start|>system\\n' + content + '<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n {%- set index = (messages|length - 1) - loop.index0 %}\n {%- if ns.multi_step_tool and message.role == \"user\" %}\n {%- set content = render_content(message.content, false)|trim %}\n {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}\n {%- set ns.multi_step_tool = false %}\n {%- set ns.last_query_index = index %}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if ns.multi_step_tool %}\n {{- raise_exception('No user query found in messages.') }}\n{%- endif %}\n{%- for message in messages %}\n {%- set content = render_content(message.content, true)|trim %}\n {%- if message.role == \"system\" %}\n {%- if not loop.first %}\n {{- raise_exception('System message must be at the beginning.') }}\n {%- endif %}\n {%- elif message.role == \"user\" %}\n {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {%- set reasoning_content = '' %}\n {%- if message.reasoning_content is string %}\n {%- set reasoning_content = message.reasoning_content %}\n {%- else %}\n {%- if '</think>' in content %}\n {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n {%- endif %}\n {%- endif %}\n {%- set reasoning_content = reasoning_content|trim %}\n {%- if loop.index0 > ns.last_query_index %}\n {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content + '\\n</think>\\n\\n' + content }}\n {%- else %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- endif %}\n {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {%- if loop.first %}\n {%- if content|trim %}\n {{- '\\n\\n<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- else %}\n {{- '<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- endif %}\n {%- else %}\n {{- '\\n<tool_call>\\n<function=' + tool_call.name + '>\\n' }}\n {%- endif %}\n {%- if tool_call.arguments is defined %}\n {%- for args_name, args_value in tool_call.arguments|items %}\n {{- '<parameter=' + args_name + '>\\n' }}\n {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}\n {{- args_value }}\n {{- '\\n</parameter>\\n' }}\n {%- endfor %}\n {%- endif %}\n {{- '</function>\\n</tool_call>' }}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.previtem and loop.previtem.role != \"tool\" %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- content }}\n {{- '\\n</tool_response>' }}\n {%- if not loop.last and loop.nextitem.role != \"tool\" %}\n {{- '<|im_end|>\\n' }}\n {%- elif loop.last %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- else %}\n {{- raise_exception('Unexpected message role.') }}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n {%- if enable_thinking is defined and enable_thinking is true %}\n {{- '<think>\\n' }}\n {%- else %}\n {{- '<think>\\n\\n</think>\\n\\n' }}\n {%- endif %}\n{%- endif %}",
286
+ "clean_up_tokenization_spaces": false,
287
+ "eos_token": "<|im_end|>",
288
+ "errors": "replace",
289
+ "model_max_length": 262144,
290
+ "pad_token": "<|endoftext|>",
291
+ "split_special_tokens": false,
292
+ "tokenizer_class": "Qwen2Tokenizer",
293
+ "unk_token": null,
294
+ "add_bos_token": false,
295
+ "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
296
+ "extra_special_tokens": {
297
+ "audio_bos_token": "<|audio_start|>",
298
+ "audio_eos_token": "<|audio_end|>",
299
+ "audio_token": "<|audio_pad|>",
300
+ "image_token": "<|image_pad|>",
301
+ "video_token": "<|video_pad|>",
302
+ "vision_bos_token": "<|vision_start|>",
303
+ "vision_eos_token": "<|vision_end|>"
304
+ }
305
+ }
v1_edit_report.json ADDED
@@ -0,0 +1,3218 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "status": "experimental_unvalidated",
3
+ "improvement_established": false,
4
+ "forward_passes": 0,
5
+ "neural_network_runtime_imported": false,
6
+ "behavioral_evaluation": "Not run, per user constraint",
7
+ "source": "Qwen/Qwen3.5-0.8B",
8
+ "revision": "2fc06364715b967f1860aea9cf38778875588b17",
9
+ "method": "Positive static vocabulary alignment; sparse 1.02x MLP output-column scaling",
10
+ "formula": "d = unit(mean(unit(domain_embedding_rows)) - mean(unit(control_embedding_rows))); score_j = dot(d, unit(down_proj[:,j])); W_new[:,j] = BF16(1.02 * FP32(W_old[:,j]))",
11
+ "selection": {
12
+ "layer_range_inclusive": [
13
+ 12,
14
+ 23
15
+ ],
16
+ "columns": 64,
17
+ "max_per_layer": 8,
18
+ "scale": 1.02
19
+ },
20
+ "domain_vocabulary": {
21
+ "ids": [
22
+ 1265,
23
+ 1528,
24
+ 1852,
25
+ 1898,
26
+ 1938,
27
+ 2374,
28
+ 2424,
29
+ 2708,
30
+ 3019,
31
+ 3166,
32
+ 3288,
33
+ 3294,
34
+ 3559,
35
+ 3885,
36
+ 4425,
37
+ 5134,
38
+ 5436,
39
+ 5614,
40
+ 5975,
41
+ 6043,
42
+ 7187,
43
+ 7450,
44
+ 8578,
45
+ 8842,
46
+ 9668,
47
+ 9714,
48
+ 10083,
49
+ 10480,
50
+ 21252,
51
+ 28619,
52
+ 31012,
53
+ 32817,
54
+ 38625,
55
+ 41078,
56
+ 46819,
57
+ 51390,
58
+ 63851,
59
+ 76706,
60
+ 77651,
61
+ 81254,
62
+ 81762
63
+ ],
64
+ "included": [
65
+ {
66
+ "text": "block",
67
+ "ids": [
68
+ 4425
69
+ ],
70
+ "tokens": [
71
+ "block"
72
+ ]
73
+ },
74
+ {
75
+ "text": " block",
76
+ "ids": [
77
+ 2424
78
+ ],
79
+ "tokens": [
80
+ "\u0120block"
81
+ ]
82
+ },
83
+ {
84
+ "text": "blocks",
85
+ "ids": [
86
+ 21252
87
+ ],
88
+ "tokens": [
89
+ "blocks"
90
+ ]
91
+ },
92
+ {
93
+ "text": " blocks",
94
+ "ids": [
95
+ 9714
96
+ ],
97
+ "tokens": [
98
+ "\u0120blocks"
99
+ ]
100
+ },
101
+ {
102
+ "text": "stack",
103
+ "ids": [
104
+ 7450
105
+ ],
106
+ "tokens": [
107
+ "stack"
108
+ ]
109
+ },
110
+ {
111
+ "text": " stack",
112
+ "ids": [
113
+ 5436
114
+ ],
115
+ "tokens": [
116
+ "\u0120stack"
117
+ ]
118
+ },
119
+ {
120
+ "text": " stacks",
121
+ "ids": [
122
+ 38625
123
+ ],
124
+ "tokens": [
125
+ "\u0120stacks"
126
+ ]
127
+ },
128
+ {
129
+ "text": "pickup",
130
+ "ids": [
131
+ 81762
132
+ ],
133
+ "tokens": [
134
+ "pickup"
135
+ ]
136
+ },
137
+ {
138
+ "text": " pickup",
139
+ "ids": [
140
+ 28619
141
+ ],
142
+ "tokens": [
143
+ "\u0120pickup"
144
+ ]
145
+ },
146
+ {
147
+ "text": "table",
148
+ "ids": [
149
+ 1938
150
+ ],
151
+ "tokens": [
152
+ "table"
153
+ ]
154
+ },
155
+ {
156
+ "text": " table",
157
+ "ids": [
158
+ 1898
159
+ ],
160
+ "tokens": [
161
+ "\u0120table"
162
+ ]
163
+ },
164
+ {
165
+ "text": "clear",
166
+ "ids": [
167
+ 7187
168
+ ],
169
+ "tokens": [
170
+ "clear"
171
+ ]
172
+ },
173
+ {
174
+ "text": " clear",
175
+ "ids": [
176
+ 2708
177
+ ],
178
+ "tokens": [
179
+ "\u0120clear"
180
+ ]
181
+ },
182
+ {
183
+ "text": "holding",
184
+ "ids": [
185
+ 41078
186
+ ],
187
+ "tokens": [
188
+ "holding"
189
+ ]
190
+ },
191
+ {
192
+ "text": " holding",
193
+ "ids": [
194
+ 9668
195
+ ],
196
+ "tokens": [
197
+ "\u0120holding"
198
+ ]
199
+ },
200
+ {
201
+ "text": "above",
202
+ "ids": [
203
+ 46819
204
+ ],
205
+ "tokens": [
206
+ "above"
207
+ ]
208
+ },
209
+ {
210
+ "text": " above",
211
+ "ids": [
212
+ 3294
213
+ ],
214
+ "tokens": [
215
+ "\u0120above"
216
+ ]
217
+ },
218
+ {
219
+ "text": "below",
220
+ "ids": [
221
+ 51390
222
+ ],
223
+ "tokens": [
224
+ "below"
225
+ ]
226
+ },
227
+ {
228
+ "text": " below",
229
+ "ids": [
230
+ 3559
231
+ ],
232
+ "tokens": [
233
+ "\u0120below"
234
+ ]
235
+ },
236
+ {
237
+ "text": "move",
238
+ "ids": [
239
+ 3288
240
+ ],
241
+ "tokens": [
242
+ "move"
243
+ ]
244
+ },
245
+ {
246
+ "text": " move",
247
+ "ids": [
248
+ 3166
249
+ ],
250
+ "tokens": [
251
+ "\u0120move"
252
+ ]
253
+ },
254
+ {
255
+ "text": "moves",
256
+ "ids": [
257
+ 63851
258
+ ],
259
+ "tokens": [
260
+ "moves"
261
+ ]
262
+ },
263
+ {
264
+ "text": " moves",
265
+ "ids": [
266
+ 10480
267
+ ],
268
+ "tokens": [
269
+ "\u0120moves"
270
+ ]
271
+ },
272
+ {
273
+ "text": "goal",
274
+ "ids": [
275
+ 32817
276
+ ],
277
+ "tokens": [
278
+ "goal"
279
+ ]
280
+ },
281
+ {
282
+ "text": " goal",
283
+ "ids": [
284
+ 5614
285
+ ],
286
+ "tokens": [
287
+ "\u0120goal"
288
+ ]
289
+ },
290
+ {
291
+ "text": "goals",
292
+ "ids": [
293
+ 81254
294
+ ],
295
+ "tokens": [
296
+ "goals"
297
+ ]
298
+ },
299
+ {
300
+ "text": " goals",
301
+ "ids": [
302
+ 8578
303
+ ],
304
+ "tokens": [
305
+ "\u0120goals"
306
+ ]
307
+ },
308
+ {
309
+ "text": "plan",
310
+ "ids": [
311
+ 10083
312
+ ],
313
+ "tokens": [
314
+ "plan"
315
+ ]
316
+ },
317
+ {
318
+ "text": " plan",
319
+ "ids": [
320
+ 3019
321
+ ],
322
+ "tokens": [
323
+ "\u0120plan"
324
+ ]
325
+ },
326
+ {
327
+ "text": " planning",
328
+ "ids": [
329
+ 8842
330
+ ],
331
+ "tokens": [
332
+ "\u0120planning"
333
+ ]
334
+ },
335
+ {
336
+ "text": " precondition",
337
+ "ids": [
338
+ 76706
339
+ ],
340
+ "tokens": [
341
+ "\u0120precondition"
342
+ ]
343
+ },
344
+ {
345
+ "text": "effects",
346
+ "ids": [
347
+ 77651
348
+ ],
349
+ "tokens": [
350
+ "effects"
351
+ ]
352
+ },
353
+ {
354
+ "text": " effects",
355
+ "ids": [
356
+ 6043
357
+ ],
358
+ "tokens": [
359
+ "\u0120effects"
360
+ ]
361
+ },
362
+ {
363
+ "text": "action",
364
+ "ids": [
365
+ 1265
366
+ ],
367
+ "tokens": [
368
+ "action"
369
+ ]
370
+ },
371
+ {
372
+ "text": " action",
373
+ "ids": [
374
+ 1852
375
+ ],
376
+ "tokens": [
377
+ "\u0120action"
378
+ ]
379
+ },
380
+ {
381
+ "text": "actions",
382
+ "ids": [
383
+ 3885
384
+ ],
385
+ "tokens": [
386
+ "actions"
387
+ ]
388
+ },
389
+ {
390
+ "text": " actions",
391
+ "ids": [
392
+ 5975
393
+ ],
394
+ "tokens": [
395
+ "\u0120actions"
396
+ ]
397
+ },
398
+ {
399
+ "text": "state",
400
+ "ids": [
401
+ 2374
402
+ ],
403
+ "tokens": [
404
+ "state"
405
+ ]
406
+ },
407
+ {
408
+ "text": " state",
409
+ "ids": [
410
+ 1528
411
+ ],
412
+ "tokens": [
413
+ "\u0120state"
414
+ ]
415
+ },
416
+ {
417
+ "text": "states",
418
+ "ids": [
419
+ 31012
420
+ ],
421
+ "tokens": [
422
+ "states"
423
+ ]
424
+ },
425
+ {
426
+ "text": " states",
427
+ "ids": [
428
+ 5134
429
+ ],
430
+ "tokens": [
431
+ "\u0120states"
432
+ ]
433
+ }
434
+ ],
435
+ "excluded_multitoken": [
436
+ {
437
+ "text": "stacks",
438
+ "ids": [
439
+ 7450,
440
+ 82
441
+ ],
442
+ "tokens": [
443
+ "stack",
444
+ "s"
445
+ ]
446
+ },
447
+ {
448
+ "text": "unstack",
449
+ "ids": [
450
+ 354,
451
+ 7450
452
+ ],
453
+ "tokens": [
454
+ "un",
455
+ "stack"
456
+ ]
457
+ },
458
+ {
459
+ "text": " unstack",
460
+ "ids": [
461
+ 632,
462
+ 7450
463
+ ],
464
+ "tokens": [
465
+ "\u0120un",
466
+ "stack"
467
+ ]
468
+ },
469
+ {
470
+ "text": "putdown",
471
+ "ids": [
472
+ 611,
473
+ 2829
474
+ ],
475
+ "tokens": [
476
+ "put",
477
+ "down"
478
+ ]
479
+ },
480
+ {
481
+ "text": " putdown",
482
+ "ids": [
483
+ 2113,
484
+ 2829
485
+ ],
486
+ "tokens": [
487
+ "\u0120put",
488
+ "down"
489
+ ]
490
+ },
491
+ {
492
+ "text": "handempty",
493
+ "ids": [
494
+ 10346,
495
+ 3092
496
+ ],
497
+ "tokens": [
498
+ "hand",
499
+ "empty"
500
+ ]
501
+ },
502
+ {
503
+ "text": " handempty",
504
+ "ids": [
505
+ 1375,
506
+ 3092
507
+ ],
508
+ "tokens": [
509
+ "\u0120hand",
510
+ "empty"
511
+ ]
512
+ },
513
+ {
514
+ "text": "ontable",
515
+ "ids": [
516
+ 530,
517
+ 470
518
+ ],
519
+ "tokens": [
520
+ "ont",
521
+ "able"
522
+ ]
523
+ },
524
+ {
525
+ "text": " ontable",
526
+ "ids": [
527
+ 383,
528
+ 1938
529
+ ],
530
+ "tokens": [
531
+ "\u0120on",
532
+ "table"
533
+ ]
534
+ },
535
+ {
536
+ "text": "planning",
537
+ "ids": [
538
+ 490,
539
+ 5762
540
+ ],
541
+ "tokens": [
542
+ "pl",
543
+ "anning"
544
+ ]
545
+ },
546
+ {
547
+ "text": "precondition",
548
+ "ids": [
549
+ 1666,
550
+ 8783
551
+ ],
552
+ "tokens": [
553
+ "pre",
554
+ "condition"
555
+ ]
556
+ },
557
+ {
558
+ "text": "preconditions",
559
+ "ids": [
560
+ 1666,
561
+ 16011
562
+ ],
563
+ "tokens": [
564
+ "pre",
565
+ "conditions"
566
+ ]
567
+ },
568
+ {
569
+ "text": " preconditions",
570
+ "ids": [
571
+ 829,
572
+ 16011
573
+ ],
574
+ "tokens": [
575
+ "\u0120pre",
576
+ "conditions"
577
+ ]
578
+ }
579
+ ]
580
+ },
581
+ "reference_vocabulary": {
582
+ "ids": [
583
+ 3487,
584
+ 4477,
585
+ 5319,
586
+ 5444,
587
+ 5522,
588
+ 8250,
589
+ 8693,
590
+ 8831,
591
+ 9228,
592
+ 9541,
593
+ 10793,
594
+ 10829,
595
+ 10849,
596
+ 10994,
597
+ 11338,
598
+ 11745,
599
+ 13159,
600
+ 13515,
601
+ 14764,
602
+ 14810,
603
+ 16666,
604
+ 16725,
605
+ 17415,
606
+ 18066,
607
+ 18255,
608
+ 18698,
609
+ 18973,
610
+ 19422,
611
+ 20512,
612
+ 21655,
613
+ 24408,
614
+ 28142,
615
+ 30141,
616
+ 30223,
617
+ 30709,
618
+ 37471,
619
+ 38115,
620
+ 38383,
621
+ 48181,
622
+ 65927,
623
+ 72497,
624
+ 80678,
625
+ 81425,
626
+ 85617,
627
+ 95456
628
+ ],
629
+ "included": [
630
+ {
631
+ "text": "music",
632
+ "ids": [
633
+ 30141
634
+ ],
635
+ "tokens": [
636
+ "music"
637
+ ]
638
+ },
639
+ {
640
+ "text": " music",
641
+ "ids": [
642
+ 4477
643
+ ],
644
+ "tokens": [
645
+ "\u0120music"
646
+ ]
647
+ },
648
+ {
649
+ "text": "song",
650
+ "ids": [
651
+ 30709
652
+ ],
653
+ "tokens": [
654
+ "song"
655
+ ]
656
+ },
657
+ {
658
+ "text": " song",
659
+ "ids": [
660
+ 5319
661
+ ],
662
+ "tokens": [
663
+ "\u0120song"
664
+ ]
665
+ },
666
+ {
667
+ "text": "songs",
668
+ "ids": [
669
+ 72497
670
+ ],
671
+ "tokens": [
672
+ "songs"
673
+ ]
674
+ },
675
+ {
676
+ "text": " songs",
677
+ "ids": [
678
+ 11338
679
+ ],
680
+ "tokens": [
681
+ "\u0120songs"
682
+ ]
683
+ },
684
+ {
685
+ "text": "movie",
686
+ "ids": [
687
+ 19422
688
+ ],
689
+ "tokens": [
690
+ "movie"
691
+ ]
692
+ },
693
+ {
694
+ "text": " movie",
695
+ "ids": [
696
+ 5522
697
+ ],
698
+ "tokens": [
699
+ "\u0120movie"
700
+ ]
701
+ },
702
+ {
703
+ "text": "movies",
704
+ "ids": [
705
+ 38115
706
+ ],
707
+ "tokens": [
708
+ "movies"
709
+ ]
710
+ },
711
+ {
712
+ "text": " movies",
713
+ "ids": [
714
+ 9228
715
+ ],
716
+ "tokens": [
717
+ "\u0120movies"
718
+ ]
719
+ },
720
+ {
721
+ "text": "actor",
722
+ "ids": [
723
+ 5444
724
+ ],
725
+ "tokens": [
726
+ "actor"
727
+ ]
728
+ },
729
+ {
730
+ "text": " actor",
731
+ "ids": [
732
+ 11745
733
+ ],
734
+ "tokens": [
735
+ "\u0120actor"
736
+ ]
737
+ },
738
+ {
739
+ "text": "actors",
740
+ "ids": [
741
+ 20512
742
+ ],
743
+ "tokens": [
744
+ "actors"
745
+ ]
746
+ },
747
+ {
748
+ "text": " actors",
749
+ "ids": [
750
+ 18973
751
+ ],
752
+ "tokens": [
753
+ "\u0120actors"
754
+ ]
755
+ },
756
+ {
757
+ "text": "recipe",
758
+ "ids": [
759
+ 24408
760
+ ],
761
+ "tokens": [
762
+ "recipe"
763
+ ]
764
+ },
765
+ {
766
+ "text": " recipe",
767
+ "ids": [
768
+ 10793
769
+ ],
770
+ "tokens": [
771
+ "\u0120recipe"
772
+ ]
773
+ },
774
+ {
775
+ "text": "recipes",
776
+ "ids": [
777
+ 48181
778
+ ],
779
+ "tokens": [
780
+ "recipes"
781
+ ]
782
+ },
783
+ {
784
+ "text": " recipes",
785
+ "ids": [
786
+ 18066
787
+ ],
788
+ "tokens": [
789
+ "\u0120recipes"
790
+ ]
791
+ },
792
+ {
793
+ "text": "food",
794
+ "ids": [
795
+ 13515
796
+ ],
797
+ "tokens": [
798
+ "food"
799
+ ]
800
+ },
801
+ {
802
+ "text": " food",
803
+ "ids": [
804
+ 3487
805
+ ],
806
+ "tokens": [
807
+ "\u0120food"
808
+ ]
809
+ },
810
+ {
811
+ "text": " cooking",
812
+ "ids": [
813
+ 16725
814
+ ],
815
+ "tokens": [
816
+ "\u0120cooking"
817
+ ]
818
+ },
819
+ {
820
+ "text": " flavor",
821
+ "ids": [
822
+ 16666
823
+ ],
824
+ "tokens": [
825
+ "\u0120flavor"
826
+ ]
827
+ },
828
+ {
829
+ "text": " fashion",
830
+ "ids": [
831
+ 10829
832
+ ],
833
+ "tokens": [
834
+ "\u0120fashion"
835
+ ]
836
+ },
837
+ {
838
+ "text": "dress",
839
+ "ids": [
840
+ 65927
841
+ ],
842
+ "tokens": [
843
+ "dress"
844
+ ]
845
+ },
846
+ {
847
+ "text": " dress",
848
+ "ids": [
849
+ 8250
850
+ ],
851
+ "tokens": [
852
+ "\u0120dress"
853
+ ]
854
+ },
855
+ {
856
+ "text": "sports",
857
+ "ids": [
858
+ 80678
859
+ ],
860
+ "tokens": [
861
+ "sports"
862
+ ]
863
+ },
864
+ {
865
+ "text": " sports",
866
+ "ids": [
867
+ 9541
868
+ ],
869
+ "tokens": [
870
+ "\u0120sports"
871
+ ]
872
+ },
873
+ {
874
+ "text": "football",
875
+ "ids": [
876
+ 81425
877
+ ],
878
+ "tokens": [
879
+ "football"
880
+ ]
881
+ },
882
+ {
883
+ "text": " football",
884
+ "ids": [
885
+ 8693
886
+ ],
887
+ "tokens": [
888
+ "\u0120football"
889
+ ]
890
+ },
891
+ {
892
+ "text": "weather",
893
+ "ids": [
894
+ 14764
895
+ ],
896
+ "tokens": [
897
+ "weather"
898
+ ]
899
+ },
900
+ {
901
+ "text": " weather",
902
+ "ids": [
903
+ 8831
904
+ ],
905
+ "tokens": [
906
+ "\u0120weather"
907
+ ]
908
+ },
909
+ {
910
+ "text": "rain",
911
+ "ids": [
912
+ 28142
913
+ ],
914
+ "tokens": [
915
+ "rain"
916
+ ]
917
+ },
918
+ {
919
+ "text": " rain",
920
+ "ids": [
921
+ 10849
922
+ ],
923
+ "tokens": [
924
+ "\u0120rain"
925
+ ]
926
+ },
927
+ {
928
+ "text": " sunny",
929
+ "ids": [
930
+ 38383
931
+ ],
932
+ "tokens": [
933
+ "\u0120sunny"
934
+ ]
935
+ },
936
+ {
937
+ "text": " beach",
938
+ "ids": [
939
+ 10994
940
+ ],
941
+ "tokens": [
942
+ "\u0120beach"
943
+ ]
944
+ },
945
+ {
946
+ "text": " ocean",
947
+ "ids": [
948
+ 17415
949
+ ],
950
+ "tokens": [
951
+ "\u0120ocean"
952
+ ]
953
+ },
954
+ {
955
+ "text": "flower",
956
+ "ids": [
957
+ 37471
958
+ ],
959
+ "tokens": [
960
+ "flower"
961
+ ]
962
+ },
963
+ {
964
+ "text": " flower",
965
+ "ids": [
966
+ 21655
967
+ ],
968
+ "tokens": [
969
+ "\u0120flower"
970
+ ]
971
+ },
972
+ {
973
+ "text": "flowers",
974
+ "ids": [
975
+ 85617
976
+ ],
977
+ "tokens": [
978
+ "flowers"
979
+ ]
980
+ },
981
+ {
982
+ "text": " flowers",
983
+ "ids": [
984
+ 18698
985
+ ],
986
+ "tokens": [
987
+ "\u0120flowers"
988
+ ]
989
+ },
990
+ {
991
+ "text": " garden",
992
+ "ids": [
993
+ 13159
994
+ ],
995
+ "tokens": [
996
+ "\u0120garden"
997
+ ]
998
+ },
999
+ {
1000
+ "text": " painting",
1001
+ "ids": [
1002
+ 18255
1003
+ ],
1004
+ "tokens": [
1005
+ "\u0120painting"
1006
+ ]
1007
+ },
1008
+ {
1009
+ "text": " poetry",
1010
+ "ids": [
1011
+ 30223
1012
+ ],
1013
+ "tokens": [
1014
+ "\u0120poetry"
1015
+ ]
1016
+ },
1017
+ {
1018
+ "text": "dance",
1019
+ "ids": [
1020
+ 95456
1021
+ ],
1022
+ "tokens": [
1023
+ "dance"
1024
+ ]
1025
+ },
1026
+ {
1027
+ "text": " dance",
1028
+ "ids": [
1029
+ 14810
1030
+ ],
1031
+ "tokens": [
1032
+ "\u0120dance"
1033
+ ]
1034
+ }
1035
+ ],
1036
+ "excluded_multitoken": [
1037
+ {
1038
+ "text": "cooking",
1039
+ "ids": [
1040
+ 981,
1041
+ 10430
1042
+ ],
1043
+ "tokens": [
1044
+ "co",
1045
+ "oking"
1046
+ ]
1047
+ },
1048
+ {
1049
+ "text": "flavor",
1050
+ "ids": [
1051
+ 1435,
1052
+ 3185
1053
+ ],
1054
+ "tokens": [
1055
+ "fl",
1056
+ "avor"
1057
+ ]
1058
+ },
1059
+ {
1060
+ "text": "fashion",
1061
+ "ids": [
1062
+ 69,
1063
+ 8517
1064
+ ],
1065
+ "tokens": [
1066
+ "f",
1067
+ "ashion"
1068
+ ]
1069
+ },
1070
+ {
1071
+ "text": "sunny",
1072
+ "ids": [
1073
+ 82,
1074
+ 26415
1075
+ ],
1076
+ "tokens": [
1077
+ "s",
1078
+ "unny"
1079
+ ]
1080
+ },
1081
+ {
1082
+ "text": "beach",
1083
+ "ids": [
1084
+ 1323,
1085
+ 595
1086
+ ],
1087
+ "tokens": [
1088
+ "be",
1089
+ "ach"
1090
+ ]
1091
+ },
1092
+ {
1093
+ "text": "ocean",
1094
+ "ids": [
1095
+ 78,
1096
+ 10880
1097
+ ],
1098
+ "tokens": [
1099
+ "o",
1100
+ "cean"
1101
+ ]
1102
+ },
1103
+ {
1104
+ "text": "garden",
1105
+ "ids": [
1106
+ 70,
1107
+ 8082
1108
+ ],
1109
+ "tokens": [
1110
+ "g",
1111
+ "arden"
1112
+ ]
1113
+ },
1114
+ {
1115
+ "text": "painting",
1116
+ "ids": [
1117
+ 32505,
1118
+ 286
1119
+ ],
1120
+ "tokens": [
1121
+ "paint",
1122
+ "ing"
1123
+ ]
1124
+ },
1125
+ {
1126
+ "text": "poetry",
1127
+ "ids": [
1128
+ 5200,
1129
+ 14707
1130
+ ],
1131
+ "tokens": [
1132
+ "po",
1133
+ "etry"
1134
+ ]
1135
+ }
1136
+ ]
1137
+ },
1138
+ "limitations": [
1139
+ "Vocabulary alignment does not identify a causal planning circuit.",
1140
+ "No evidence that selected neurons activate on Blocksworld inputs.",
1141
+ "Gated MLP activations can have either sign; amplification need not favor these terms.",
1142
+ "Residual representations and subsequent layers can change the effect.",
1143
+ "Terms are polysemous; common planning words are not specific to Blocksworld.",
1144
+ "This edit does not encode action preconditions, state transitions, or a search algorithm.",
1145
+ "Layer range, vocabulary, column count, and scale were hand-chosen, not tuned.",
1146
+ "Blocksworld performance and general capabilities may improve, stay unchanged, or regress."
1147
+ ],
1148
+ "edits": [
1149
+ {
1150
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1151
+ "layer": 23,
1152
+ "column": 316,
1153
+ "static_cosine": 0.22491052746772766,
1154
+ "changed_scalars": 1024,
1155
+ "relative_l2_change": 0.020385827869176865
1156
+ },
1157
+ {
1158
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1159
+ "layer": 21,
1160
+ "column": 3489,
1161
+ "static_cosine": 0.21439218521118164,
1162
+ "changed_scalars": 1024,
1163
+ "relative_l2_change": 0.02034873142838478
1164
+ },
1165
+ {
1166
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1167
+ "layer": 23,
1168
+ "column": 1346,
1169
+ "static_cosine": 0.20872575044631958,
1170
+ "changed_scalars": 1024,
1171
+ "relative_l2_change": 0.020354080945253372
1172
+ },
1173
+ {
1174
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1175
+ "layer": 23,
1176
+ "column": 659,
1177
+ "static_cosine": 0.20740415155887604,
1178
+ "changed_scalars": 1024,
1179
+ "relative_l2_change": 0.02033684216439724
1180
+ },
1181
+ {
1182
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1183
+ "layer": 23,
1184
+ "column": 128,
1185
+ "static_cosine": 0.18711930513381958,
1186
+ "changed_scalars": 1024,
1187
+ "relative_l2_change": 0.020345721393823624
1188
+ },
1189
+ {
1190
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1191
+ "layer": 23,
1192
+ "column": 1385,
1193
+ "static_cosine": 0.18549516797065735,
1194
+ "changed_scalars": 1024,
1195
+ "relative_l2_change": 0.020347312092781067
1196
+ },
1197
+ {
1198
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1199
+ "layer": 23,
1200
+ "column": 1323,
1201
+ "static_cosine": 0.18127796053886414,
1202
+ "changed_scalars": 1024,
1203
+ "relative_l2_change": 0.020501380786299706
1204
+ },
1205
+ {
1206
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1207
+ "layer": 23,
1208
+ "column": 368,
1209
+ "static_cosine": 0.17719918489456177,
1210
+ "changed_scalars": 1024,
1211
+ "relative_l2_change": 0.02032538130879402
1212
+ },
1213
+ {
1214
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1215
+ "layer": 22,
1216
+ "column": 186,
1217
+ "static_cosine": 0.17397309839725494,
1218
+ "changed_scalars": 1024,
1219
+ "relative_l2_change": 0.02045033499598503
1220
+ },
1221
+ {
1222
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1223
+ "layer": 23,
1224
+ "column": 834,
1225
+ "static_cosine": 0.1738962084054947,
1226
+ "changed_scalars": 1024,
1227
+ "relative_l2_change": 0.02036205120384693
1228
+ },
1229
+ {
1230
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
1231
+ "layer": 17,
1232
+ "column": 1016,
1233
+ "static_cosine": 0.1622765213251114,
1234
+ "changed_scalars": 1024,
1235
+ "relative_l2_change": 0.020449388772249222
1236
+ },
1237
+ {
1238
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1239
+ "layer": 21,
1240
+ "column": 560,
1241
+ "static_cosine": 0.1531076580286026,
1242
+ "changed_scalars": 1024,
1243
+ "relative_l2_change": 0.020451636984944344
1244
+ },
1245
+ {
1246
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1247
+ "layer": 22,
1248
+ "column": 1723,
1249
+ "static_cosine": 0.15255765616893768,
1250
+ "changed_scalars": 1024,
1251
+ "relative_l2_change": 0.020314600318670273
1252
+ },
1253
+ {
1254
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1255
+ "layer": 19,
1256
+ "column": 374,
1257
+ "static_cosine": 0.14157021045684814,
1258
+ "changed_scalars": 1024,
1259
+ "relative_l2_change": 0.020385051146149635
1260
+ },
1261
+ {
1262
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1263
+ "layer": 16,
1264
+ "column": 2700,
1265
+ "static_cosine": 0.13809005916118622,
1266
+ "changed_scalars": 1024,
1267
+ "relative_l2_change": 0.02038157917559147
1268
+ },
1269
+ {
1270
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1271
+ "layer": 21,
1272
+ "column": 845,
1273
+ "static_cosine": 0.13749338686466217,
1274
+ "changed_scalars": 1024,
1275
+ "relative_l2_change": 0.020315641537308693
1276
+ },
1277
+ {
1278
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
1279
+ "layer": 18,
1280
+ "column": 1536,
1281
+ "static_cosine": 0.13671903312206268,
1282
+ "changed_scalars": 1024,
1283
+ "relative_l2_change": 0.02040751278400421
1284
+ },
1285
+ {
1286
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
1287
+ "layer": 20,
1288
+ "column": 1106,
1289
+ "static_cosine": 0.13618186116218567,
1290
+ "changed_scalars": 1024,
1291
+ "relative_l2_change": 0.020227784290909767
1292
+ },
1293
+ {
1294
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1295
+ "layer": 21,
1296
+ "column": 1187,
1297
+ "static_cosine": 0.13517242670059204,
1298
+ "changed_scalars": 1024,
1299
+ "relative_l2_change": 0.02042551338672638
1300
+ },
1301
+ {
1302
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1303
+ "layer": 22,
1304
+ "column": 367,
1305
+ "static_cosine": 0.13117794692516327,
1306
+ "changed_scalars": 1024,
1307
+ "relative_l2_change": 0.020488491281867027
1308
+ },
1309
+ {
1310
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
1311
+ "layer": 20,
1312
+ "column": 651,
1313
+ "static_cosine": 0.13094821572303772,
1314
+ "changed_scalars": 1024,
1315
+ "relative_l2_change": 0.020474741235375404
1316
+ },
1317
+ {
1318
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1319
+ "layer": 21,
1320
+ "column": 3333,
1321
+ "static_cosine": 0.1308772712945938,
1322
+ "changed_scalars": 1024,
1323
+ "relative_l2_change": 0.02036348171532154
1324
+ },
1325
+ {
1326
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
1327
+ "layer": 15,
1328
+ "column": 3319,
1329
+ "static_cosine": 0.12753400206565857,
1330
+ "changed_scalars": 1024,
1331
+ "relative_l2_change": 0.02063577249646187
1332
+ },
1333
+ {
1334
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
1335
+ "layer": 20,
1336
+ "column": 205,
1337
+ "static_cosine": 0.1268293708562851,
1338
+ "changed_scalars": 1024,
1339
+ "relative_l2_change": 0.02038157731294632
1340
+ },
1341
+ {
1342
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
1343
+ "layer": 14,
1344
+ "column": 1426,
1345
+ "static_cosine": 0.12495019286870956,
1346
+ "changed_scalars": 1024,
1347
+ "relative_l2_change": 0.020464153960347176
1348
+ },
1349
+ {
1350
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1351
+ "layer": 19,
1352
+ "column": 2782,
1353
+ "static_cosine": 0.12428195774555206,
1354
+ "changed_scalars": 1024,
1355
+ "relative_l2_change": 0.020458294078707695
1356
+ },
1357
+ {
1358
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1359
+ "layer": 19,
1360
+ "column": 3417,
1361
+ "static_cosine": 0.11998037993907928,
1362
+ "changed_scalars": 1024,
1363
+ "relative_l2_change": 0.0204093549400568
1364
+ },
1365
+ {
1366
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1367
+ "layer": 22,
1368
+ "column": 2286,
1369
+ "static_cosine": 0.11944878101348877,
1370
+ "changed_scalars": 1024,
1371
+ "relative_l2_change": 0.020437324419617653
1372
+ },
1373
+ {
1374
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
1375
+ "layer": 18,
1376
+ "column": 3101,
1377
+ "static_cosine": 0.11745330691337585,
1378
+ "changed_scalars": 1024,
1379
+ "relative_l2_change": 0.02034442126750946
1380
+ },
1381
+ {
1382
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
1383
+ "layer": 20,
1384
+ "column": 418,
1385
+ "static_cosine": 0.11588569730520248,
1386
+ "changed_scalars": 1024,
1387
+ "relative_l2_change": 0.02047627605497837
1388
+ },
1389
+ {
1390
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
1391
+ "layer": 17,
1392
+ "column": 377,
1393
+ "static_cosine": 0.11445152014493942,
1394
+ "changed_scalars": 1024,
1395
+ "relative_l2_change": 0.020521720871329308
1396
+ },
1397
+ {
1398
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
1399
+ "layer": 12,
1400
+ "column": 2888,
1401
+ "static_cosine": 0.11340443044900894,
1402
+ "changed_scalars": 1024,
1403
+ "relative_l2_change": 0.02035796083509922
1404
+ },
1405
+ {
1406
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1407
+ "layer": 19,
1408
+ "column": 955,
1409
+ "static_cosine": 0.11207542568445206,
1410
+ "changed_scalars": 1024,
1411
+ "relative_l2_change": 0.02025194838643074
1412
+ },
1413
+ {
1414
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1415
+ "layer": 21,
1416
+ "column": 280,
1417
+ "static_cosine": 0.1120038703083992,
1418
+ "changed_scalars": 1024,
1419
+ "relative_l2_change": 0.020400062203407288
1420
+ },
1421
+ {
1422
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
1423
+ "layer": 15,
1424
+ "column": 2922,
1425
+ "static_cosine": 0.11180732399225235,
1426
+ "changed_scalars": 1024,
1427
+ "relative_l2_change": 0.020490530878305435
1428
+ },
1429
+ {
1430
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1431
+ "layer": 22,
1432
+ "column": 1039,
1433
+ "static_cosine": 0.11040688306093216,
1434
+ "changed_scalars": 1024,
1435
+ "relative_l2_change": 0.02035714127123356
1436
+ },
1437
+ {
1438
+ "tensor": "model.language_model.layers.13.mlp.down_proj.weight",
1439
+ "layer": 13,
1440
+ "column": 2897,
1441
+ "static_cosine": 0.11006982624530792,
1442
+ "changed_scalars": 1024,
1443
+ "relative_l2_change": 0.020316656678915024
1444
+ },
1445
+ {
1446
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1447
+ "layer": 21,
1448
+ "column": 2401,
1449
+ "static_cosine": 0.10850067436695099,
1450
+ "changed_scalars": 1024,
1451
+ "relative_l2_change": 0.020381048321723938
1452
+ },
1453
+ {
1454
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
1455
+ "layer": 15,
1456
+ "column": 693,
1457
+ "static_cosine": 0.1084042638540268,
1458
+ "changed_scalars": 1024,
1459
+ "relative_l2_change": 0.02036408707499504
1460
+ },
1461
+ {
1462
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1463
+ "layer": 19,
1464
+ "column": 22,
1465
+ "static_cosine": 0.1081288531422615,
1466
+ "changed_scalars": 1024,
1467
+ "relative_l2_change": 0.020258167758584023
1468
+ },
1469
+ {
1470
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1471
+ "layer": 16,
1472
+ "column": 1971,
1473
+ "static_cosine": 0.10762576013803482,
1474
+ "changed_scalars": 1024,
1475
+ "relative_l2_change": 0.020540636032819748
1476
+ },
1477
+ {
1478
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
1479
+ "layer": 15,
1480
+ "column": 1243,
1481
+ "static_cosine": 0.10739479959011078,
1482
+ "changed_scalars": 1024,
1483
+ "relative_l2_change": 0.020336318761110306
1484
+ },
1485
+ {
1486
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1487
+ "layer": 21,
1488
+ "column": 361,
1489
+ "static_cosine": 0.10721300542354584,
1490
+ "changed_scalars": 1024,
1491
+ "relative_l2_change": 0.020364515483379364
1492
+ },
1493
+ {
1494
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
1495
+ "layer": 12,
1496
+ "column": 329,
1497
+ "static_cosine": 0.10571350902318954,
1498
+ "changed_scalars": 1024,
1499
+ "relative_l2_change": 0.020360711961984634
1500
+ },
1501
+ {
1502
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
1503
+ "layer": 15,
1504
+ "column": 731,
1505
+ "static_cosine": 0.10522869229316711,
1506
+ "changed_scalars": 1024,
1507
+ "relative_l2_change": 0.02036682330071926
1508
+ },
1509
+ {
1510
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
1511
+ "layer": 17,
1512
+ "column": 1982,
1513
+ "static_cosine": 0.10521752387285233,
1514
+ "changed_scalars": 1024,
1515
+ "relative_l2_change": 0.020364902913570404
1516
+ },
1517
+ {
1518
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1519
+ "layer": 16,
1520
+ "column": 1233,
1521
+ "static_cosine": 0.10503878444433212,
1522
+ "changed_scalars": 1024,
1523
+ "relative_l2_change": 0.020587746053934097
1524
+ },
1525
+ {
1526
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1527
+ "layer": 22,
1528
+ "column": 2496,
1529
+ "static_cosine": 0.10480718314647675,
1530
+ "changed_scalars": 1024,
1531
+ "relative_l2_change": 0.02032502368092537
1532
+ },
1533
+ {
1534
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
1535
+ "layer": 17,
1536
+ "column": 427,
1537
+ "static_cosine": 0.104188472032547,
1538
+ "changed_scalars": 1024,
1539
+ "relative_l2_change": 0.020468559116125107
1540
+ },
1541
+ {
1542
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1543
+ "layer": 22,
1544
+ "column": 690,
1545
+ "static_cosine": 0.10351220518350601,
1546
+ "changed_scalars": 1024,
1547
+ "relative_l2_change": 0.02038443274796009
1548
+ },
1549
+ {
1550
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1551
+ "layer": 19,
1552
+ "column": 2020,
1553
+ "static_cosine": 0.10346139967441559,
1554
+ "changed_scalars": 1024,
1555
+ "relative_l2_change": 0.020326655358076096
1556
+ },
1557
+ {
1558
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1559
+ "layer": 22,
1560
+ "column": 538,
1561
+ "static_cosine": 0.10315290093421936,
1562
+ "changed_scalars": 1024,
1563
+ "relative_l2_change": 0.020336532965302467
1564
+ },
1565
+ {
1566
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1567
+ "layer": 19,
1568
+ "column": 2616,
1569
+ "static_cosine": 0.10307393968105316,
1570
+ "changed_scalars": 1024,
1571
+ "relative_l2_change": 0.020424814894795418
1572
+ },
1573
+ {
1574
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1575
+ "layer": 16,
1576
+ "column": 3402,
1577
+ "static_cosine": 0.10285601764917374,
1578
+ "changed_scalars": 1024,
1579
+ "relative_l2_change": 0.020510397851467133
1580
+ },
1581
+ {
1582
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1583
+ "layer": 16,
1584
+ "column": 1570,
1585
+ "static_cosine": 0.10237244516611099,
1586
+ "changed_scalars": 1024,
1587
+ "relative_l2_change": 0.020438753068447113
1588
+ },
1589
+ {
1590
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
1591
+ "layer": 15,
1592
+ "column": 1452,
1593
+ "static_cosine": 0.10189658403396606,
1594
+ "changed_scalars": 1024,
1595
+ "relative_l2_change": 0.020365500822663307
1596
+ },
1597
+ {
1598
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
1599
+ "layer": 20,
1600
+ "column": 687,
1601
+ "static_cosine": 0.10100101679563522,
1602
+ "changed_scalars": 1024,
1603
+ "relative_l2_change": 0.02036142908036709
1604
+ },
1605
+ {
1606
+ "tensor": "model.language_model.layers.13.mlp.down_proj.weight",
1607
+ "layer": 13,
1608
+ "column": 1864,
1609
+ "static_cosine": 0.10049968212842941,
1610
+ "changed_scalars": 1024,
1611
+ "relative_l2_change": 0.020534183830022812
1612
+ },
1613
+ {
1614
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
1615
+ "layer": 15,
1616
+ "column": 3084,
1617
+ "static_cosine": 0.10034879297018051,
1618
+ "changed_scalars": 1024,
1619
+ "relative_l2_change": 0.020527463406324387
1620
+ },
1621
+ {
1622
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1623
+ "layer": 19,
1624
+ "column": 1414,
1625
+ "static_cosine": 0.10003922879695892,
1626
+ "changed_scalars": 1024,
1627
+ "relative_l2_change": 0.02048167772591114
1628
+ },
1629
+ {
1630
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
1631
+ "layer": 17,
1632
+ "column": 2892,
1633
+ "static_cosine": 0.0996561273932457,
1634
+ "changed_scalars": 1024,
1635
+ "relative_l2_change": 0.020417578518390656
1636
+ },
1637
+ {
1638
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
1639
+ "layer": 20,
1640
+ "column": 2261,
1641
+ "static_cosine": 0.099310003221035,
1642
+ "changed_scalars": 1024,
1643
+ "relative_l2_change": 0.020274925976991653
1644
+ },
1645
+ {
1646
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
1647
+ "layer": 14,
1648
+ "column": 1414,
1649
+ "static_cosine": 0.09902976453304291,
1650
+ "changed_scalars": 1024,
1651
+ "relative_l2_change": 0.02043342776596546
1652
+ },
1653
+ {
1654
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1655
+ "layer": 16,
1656
+ "column": 631,
1657
+ "static_cosine": 0.09897439181804657,
1658
+ "changed_scalars": 1024,
1659
+ "relative_l2_change": 0.020490428432822227
1660
+ }
1661
+ ],
1662
+ "top_candidates": [
1663
+ {
1664
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1665
+ "layer": 23,
1666
+ "column": 316,
1667
+ "static_cosine": 0.22491052746772766
1668
+ },
1669
+ {
1670
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1671
+ "layer": 21,
1672
+ "column": 3489,
1673
+ "static_cosine": 0.21439218521118164
1674
+ },
1675
+ {
1676
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1677
+ "layer": 23,
1678
+ "column": 1346,
1679
+ "static_cosine": 0.20872575044631958
1680
+ },
1681
+ {
1682
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1683
+ "layer": 23,
1684
+ "column": 659,
1685
+ "static_cosine": 0.20740415155887604
1686
+ },
1687
+ {
1688
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1689
+ "layer": 23,
1690
+ "column": 128,
1691
+ "static_cosine": 0.18711930513381958
1692
+ },
1693
+ {
1694
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1695
+ "layer": 23,
1696
+ "column": 1385,
1697
+ "static_cosine": 0.18549516797065735
1698
+ },
1699
+ {
1700
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1701
+ "layer": 23,
1702
+ "column": 1323,
1703
+ "static_cosine": 0.18127796053886414
1704
+ },
1705
+ {
1706
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1707
+ "layer": 23,
1708
+ "column": 368,
1709
+ "static_cosine": 0.17719918489456177
1710
+ },
1711
+ {
1712
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1713
+ "layer": 22,
1714
+ "column": 186,
1715
+ "static_cosine": 0.17397309839725494
1716
+ },
1717
+ {
1718
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1719
+ "layer": 23,
1720
+ "column": 834,
1721
+ "static_cosine": 0.1738962084054947
1722
+ },
1723
+ {
1724
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
1725
+ "layer": 17,
1726
+ "column": 1016,
1727
+ "static_cosine": 0.1622765213251114
1728
+ },
1729
+ {
1730
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1731
+ "layer": 23,
1732
+ "column": 526,
1733
+ "static_cosine": 0.1560610979795456
1734
+ },
1735
+ {
1736
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1737
+ "layer": 21,
1738
+ "column": 560,
1739
+ "static_cosine": 0.1531076580286026
1740
+ },
1741
+ {
1742
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1743
+ "layer": 22,
1744
+ "column": 1723,
1745
+ "static_cosine": 0.15255765616893768
1746
+ },
1747
+ {
1748
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1749
+ "layer": 23,
1750
+ "column": 302,
1751
+ "static_cosine": 0.15250621736049652
1752
+ },
1753
+ {
1754
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1755
+ "layer": 23,
1756
+ "column": 2697,
1757
+ "static_cosine": 0.1517026275396347
1758
+ },
1759
+ {
1760
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1761
+ "layer": 23,
1762
+ "column": 1752,
1763
+ "static_cosine": 0.14557033777236938
1764
+ },
1765
+ {
1766
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1767
+ "layer": 19,
1768
+ "column": 374,
1769
+ "static_cosine": 0.14157021045684814
1770
+ },
1771
+ {
1772
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1773
+ "layer": 16,
1774
+ "column": 2700,
1775
+ "static_cosine": 0.13809005916118622
1776
+ },
1777
+ {
1778
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1779
+ "layer": 23,
1780
+ "column": 2867,
1781
+ "static_cosine": 0.13762426376342773
1782
+ },
1783
+ {
1784
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1785
+ "layer": 21,
1786
+ "column": 845,
1787
+ "static_cosine": 0.13749338686466217
1788
+ },
1789
+ {
1790
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
1791
+ "layer": 18,
1792
+ "column": 1536,
1793
+ "static_cosine": 0.13671903312206268
1794
+ },
1795
+ {
1796
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
1797
+ "layer": 20,
1798
+ "column": 1106,
1799
+ "static_cosine": 0.13618186116218567
1800
+ },
1801
+ {
1802
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1803
+ "layer": 21,
1804
+ "column": 1187,
1805
+ "static_cosine": 0.13517242670059204
1806
+ },
1807
+ {
1808
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1809
+ "layer": 22,
1810
+ "column": 367,
1811
+ "static_cosine": 0.13117794692516327
1812
+ },
1813
+ {
1814
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
1815
+ "layer": 20,
1816
+ "column": 651,
1817
+ "static_cosine": 0.13094821572303772
1818
+ },
1819
+ {
1820
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1821
+ "layer": 21,
1822
+ "column": 3333,
1823
+ "static_cosine": 0.1308772712945938
1824
+ },
1825
+ {
1826
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1827
+ "layer": 23,
1828
+ "column": 1162,
1829
+ "static_cosine": 0.12827275693416595
1830
+ },
1831
+ {
1832
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
1833
+ "layer": 15,
1834
+ "column": 3319,
1835
+ "static_cosine": 0.12753400206565857
1836
+ },
1837
+ {
1838
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
1839
+ "layer": 20,
1840
+ "column": 205,
1841
+ "static_cosine": 0.1268293708562851
1842
+ },
1843
+ {
1844
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
1845
+ "layer": 14,
1846
+ "column": 1426,
1847
+ "static_cosine": 0.12495019286870956
1848
+ },
1849
+ {
1850
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1851
+ "layer": 19,
1852
+ "column": 2782,
1853
+ "static_cosine": 0.12428195774555206
1854
+ },
1855
+ {
1856
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1857
+ "layer": 19,
1858
+ "column": 3417,
1859
+ "static_cosine": 0.11998037993907928
1860
+ },
1861
+ {
1862
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1863
+ "layer": 22,
1864
+ "column": 2286,
1865
+ "static_cosine": 0.11944878101348877
1866
+ },
1867
+ {
1868
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
1869
+ "layer": 18,
1870
+ "column": 3101,
1871
+ "static_cosine": 0.11745330691337585
1872
+ },
1873
+ {
1874
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1875
+ "layer": 23,
1876
+ "column": 204,
1877
+ "static_cosine": 0.11742853373289108
1878
+ },
1879
+ {
1880
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1881
+ "layer": 23,
1882
+ "column": 2942,
1883
+ "static_cosine": 0.11619303375482559
1884
+ },
1885
+ {
1886
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
1887
+ "layer": 20,
1888
+ "column": 418,
1889
+ "static_cosine": 0.11588569730520248
1890
+ },
1891
+ {
1892
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1893
+ "layer": 23,
1894
+ "column": 2935,
1895
+ "static_cosine": 0.1150318905711174
1896
+ },
1897
+ {
1898
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
1899
+ "layer": 17,
1900
+ "column": 377,
1901
+ "static_cosine": 0.11445152014493942
1902
+ },
1903
+ {
1904
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1905
+ "layer": 23,
1906
+ "column": 3348,
1907
+ "static_cosine": 0.11403235793113708
1908
+ },
1909
+ {
1910
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
1911
+ "layer": 12,
1912
+ "column": 2888,
1913
+ "static_cosine": 0.11340443044900894
1914
+ },
1915
+ {
1916
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1917
+ "layer": 19,
1918
+ "column": 955,
1919
+ "static_cosine": 0.11207542568445206
1920
+ },
1921
+ {
1922
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1923
+ "layer": 21,
1924
+ "column": 280,
1925
+ "static_cosine": 0.1120038703083992
1926
+ },
1927
+ {
1928
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
1929
+ "layer": 15,
1930
+ "column": 2922,
1931
+ "static_cosine": 0.11180732399225235
1932
+ },
1933
+ {
1934
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
1935
+ "layer": 22,
1936
+ "column": 1039,
1937
+ "static_cosine": 0.11040688306093216
1938
+ },
1939
+ {
1940
+ "tensor": "model.language_model.layers.13.mlp.down_proj.weight",
1941
+ "layer": 13,
1942
+ "column": 2897,
1943
+ "static_cosine": 0.11006982624530792
1944
+ },
1945
+ {
1946
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1947
+ "layer": 21,
1948
+ "column": 2401,
1949
+ "static_cosine": 0.10850067436695099
1950
+ },
1951
+ {
1952
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
1953
+ "layer": 15,
1954
+ "column": 693,
1955
+ "static_cosine": 0.1084042638540268
1956
+ },
1957
+ {
1958
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
1959
+ "layer": 19,
1960
+ "column": 22,
1961
+ "static_cosine": 0.1081288531422615
1962
+ },
1963
+ {
1964
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
1965
+ "layer": 16,
1966
+ "column": 1971,
1967
+ "static_cosine": 0.10762576013803482
1968
+ },
1969
+ {
1970
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
1971
+ "layer": 15,
1972
+ "column": 1243,
1973
+ "static_cosine": 0.10739479959011078
1974
+ },
1975
+ {
1976
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
1977
+ "layer": 21,
1978
+ "column": 361,
1979
+ "static_cosine": 0.10721300542354584
1980
+ },
1981
+ {
1982
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1983
+ "layer": 23,
1984
+ "column": 717,
1985
+ "static_cosine": 0.1066194623708725
1986
+ },
1987
+ {
1988
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
1989
+ "layer": 23,
1990
+ "column": 2057,
1991
+ "static_cosine": 0.10609494149684906
1992
+ },
1993
+ {
1994
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
1995
+ "layer": 12,
1996
+ "column": 329,
1997
+ "static_cosine": 0.10571350902318954
1998
+ },
1999
+ {
2000
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2001
+ "layer": 23,
2002
+ "column": 3468,
2003
+ "static_cosine": 0.10562452673912048
2004
+ },
2005
+ {
2006
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
2007
+ "layer": 15,
2008
+ "column": 731,
2009
+ "static_cosine": 0.10522869229316711
2010
+ },
2011
+ {
2012
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2013
+ "layer": 17,
2014
+ "column": 1982,
2015
+ "static_cosine": 0.10521752387285233
2016
+ },
2017
+ {
2018
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2019
+ "layer": 16,
2020
+ "column": 1233,
2021
+ "static_cosine": 0.10503878444433212
2022
+ },
2023
+ {
2024
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2025
+ "layer": 22,
2026
+ "column": 2496,
2027
+ "static_cosine": 0.10480718314647675
2028
+ },
2029
+ {
2030
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2031
+ "layer": 17,
2032
+ "column": 427,
2033
+ "static_cosine": 0.104188472032547
2034
+ },
2035
+ {
2036
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2037
+ "layer": 22,
2038
+ "column": 690,
2039
+ "static_cosine": 0.10351220518350601
2040
+ },
2041
+ {
2042
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2043
+ "layer": 19,
2044
+ "column": 2020,
2045
+ "static_cosine": 0.10346139967441559
2046
+ },
2047
+ {
2048
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2049
+ "layer": 22,
2050
+ "column": 538,
2051
+ "static_cosine": 0.10315290093421936
2052
+ },
2053
+ {
2054
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2055
+ "layer": 19,
2056
+ "column": 2616,
2057
+ "static_cosine": 0.10307393968105316
2058
+ },
2059
+ {
2060
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2061
+ "layer": 16,
2062
+ "column": 3402,
2063
+ "static_cosine": 0.10285601764917374
2064
+ },
2065
+ {
2066
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2067
+ "layer": 16,
2068
+ "column": 1570,
2069
+ "static_cosine": 0.10237244516611099
2070
+ },
2071
+ {
2072
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2073
+ "layer": 22,
2074
+ "column": 2386,
2075
+ "static_cosine": 0.10213257372379303
2076
+ },
2077
+ {
2078
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
2079
+ "layer": 15,
2080
+ "column": 1452,
2081
+ "static_cosine": 0.10189658403396606
2082
+ },
2083
+ {
2084
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2085
+ "layer": 21,
2086
+ "column": 2312,
2087
+ "static_cosine": 0.10153216868638992
2088
+ },
2089
+ {
2090
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2091
+ "layer": 21,
2092
+ "column": 483,
2093
+ "static_cosine": 0.10108119249343872
2094
+ },
2095
+ {
2096
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
2097
+ "layer": 20,
2098
+ "column": 687,
2099
+ "static_cosine": 0.10100101679563522
2100
+ },
2101
+ {
2102
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2103
+ "layer": 23,
2104
+ "column": 2450,
2105
+ "static_cosine": 0.10078106820583344
2106
+ },
2107
+ {
2108
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2109
+ "layer": 23,
2110
+ "column": 1304,
2111
+ "static_cosine": 0.10068120062351227
2112
+ },
2113
+ {
2114
+ "tensor": "model.language_model.layers.13.mlp.down_proj.weight",
2115
+ "layer": 13,
2116
+ "column": 1864,
2117
+ "static_cosine": 0.10049968212842941
2118
+ },
2119
+ {
2120
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2121
+ "layer": 22,
2122
+ "column": 1105,
2123
+ "static_cosine": 0.1003764346241951
2124
+ },
2125
+ {
2126
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
2127
+ "layer": 15,
2128
+ "column": 3084,
2129
+ "static_cosine": 0.10034879297018051
2130
+ },
2131
+ {
2132
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2133
+ "layer": 21,
2134
+ "column": 2533,
2135
+ "static_cosine": 0.10016046464443207
2136
+ },
2137
+ {
2138
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2139
+ "layer": 19,
2140
+ "column": 1414,
2141
+ "static_cosine": 0.10003922879695892
2142
+ },
2143
+ {
2144
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2145
+ "layer": 23,
2146
+ "column": 890,
2147
+ "static_cosine": 0.09985574334859848
2148
+ },
2149
+ {
2150
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2151
+ "layer": 22,
2152
+ "column": 59,
2153
+ "static_cosine": 0.09968943893909454
2154
+ },
2155
+ {
2156
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2157
+ "layer": 17,
2158
+ "column": 2892,
2159
+ "static_cosine": 0.0996561273932457
2160
+ },
2161
+ {
2162
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
2163
+ "layer": 20,
2164
+ "column": 2261,
2165
+ "static_cosine": 0.099310003221035
2166
+ },
2167
+ {
2168
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2169
+ "layer": 22,
2170
+ "column": 419,
2171
+ "static_cosine": 0.09924012422561646
2172
+ },
2173
+ {
2174
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2175
+ "layer": 21,
2176
+ "column": 1001,
2177
+ "static_cosine": 0.09922237694263458
2178
+ },
2179
+ {
2180
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
2181
+ "layer": 14,
2182
+ "column": 1414,
2183
+ "static_cosine": 0.09902976453304291
2184
+ },
2185
+ {
2186
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2187
+ "layer": 16,
2188
+ "column": 631,
2189
+ "static_cosine": 0.09897439181804657
2190
+ },
2191
+ {
2192
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
2193
+ "layer": 18,
2194
+ "column": 2594,
2195
+ "static_cosine": 0.09881844371557236
2196
+ },
2197
+ {
2198
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2199
+ "layer": 17,
2200
+ "column": 2029,
2201
+ "static_cosine": 0.09840156137943268
2202
+ },
2203
+ {
2204
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
2205
+ "layer": 15,
2206
+ "column": 1297,
2207
+ "static_cosine": 0.09739521890878677
2208
+ },
2209
+ {
2210
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2211
+ "layer": 22,
2212
+ "column": 3492,
2213
+ "static_cosine": 0.09724452346563339
2214
+ },
2215
+ {
2216
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
2217
+ "layer": 18,
2218
+ "column": 2357,
2219
+ "static_cosine": 0.09716974943876266
2220
+ },
2221
+ {
2222
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2223
+ "layer": 17,
2224
+ "column": 3569,
2225
+ "static_cosine": 0.09714695811271667
2226
+ },
2227
+ {
2228
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2229
+ "layer": 16,
2230
+ "column": 2677,
2231
+ "static_cosine": 0.09712649881839752
2232
+ },
2233
+ {
2234
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2235
+ "layer": 23,
2236
+ "column": 2315,
2237
+ "static_cosine": 0.09693803638219833
2238
+ },
2239
+ {
2240
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2241
+ "layer": 19,
2242
+ "column": 3180,
2243
+ "static_cosine": 0.09678968042135239
2244
+ },
2245
+ {
2246
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
2247
+ "layer": 20,
2248
+ "column": 2324,
2249
+ "static_cosine": 0.09665478765964508
2250
+ },
2251
+ {
2252
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
2253
+ "layer": 12,
2254
+ "column": 407,
2255
+ "static_cosine": 0.09648769348859787
2256
+ },
2257
+ {
2258
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
2259
+ "layer": 15,
2260
+ "column": 1715,
2261
+ "static_cosine": 0.09642699360847473
2262
+ },
2263
+ {
2264
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2265
+ "layer": 19,
2266
+ "column": 2606,
2267
+ "static_cosine": 0.0960790142416954
2268
+ },
2269
+ {
2270
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2271
+ "layer": 23,
2272
+ "column": 1760,
2273
+ "static_cosine": 0.09592436254024506
2274
+ },
2275
+ {
2276
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
2277
+ "layer": 14,
2278
+ "column": 3330,
2279
+ "static_cosine": 0.09589985758066177
2280
+ },
2281
+ {
2282
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
2283
+ "layer": 20,
2284
+ "column": 888,
2285
+ "static_cosine": 0.09586537629365921
2286
+ },
2287
+ {
2288
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
2289
+ "layer": 12,
2290
+ "column": 1351,
2291
+ "static_cosine": 0.09585133194923401
2292
+ },
2293
+ {
2294
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2295
+ "layer": 22,
2296
+ "column": 2435,
2297
+ "static_cosine": 0.09569843858480453
2298
+ },
2299
+ {
2300
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2301
+ "layer": 17,
2302
+ "column": 2000,
2303
+ "static_cosine": 0.09567302465438843
2304
+ },
2305
+ {
2306
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
2307
+ "layer": 18,
2308
+ "column": 1811,
2309
+ "static_cosine": 0.09534972906112671
2310
+ },
2311
+ {
2312
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2313
+ "layer": 23,
2314
+ "column": 568,
2315
+ "static_cosine": 0.09506271779537201
2316
+ },
2317
+ {
2318
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2319
+ "layer": 21,
2320
+ "column": 852,
2321
+ "static_cosine": 0.09504982084035873
2322
+ },
2323
+ {
2324
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2325
+ "layer": 19,
2326
+ "column": 2583,
2327
+ "static_cosine": 0.09439187496900558
2328
+ },
2329
+ {
2330
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2331
+ "layer": 23,
2332
+ "column": 2374,
2333
+ "static_cosine": 0.09438198804855347
2334
+ },
2335
+ {
2336
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2337
+ "layer": 16,
2338
+ "column": 2602,
2339
+ "static_cosine": 0.09400500357151031
2340
+ },
2341
+ {
2342
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2343
+ "layer": 21,
2344
+ "column": 617,
2345
+ "static_cosine": 0.09390980005264282
2346
+ },
2347
+ {
2348
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2349
+ "layer": 21,
2350
+ "column": 72,
2351
+ "static_cosine": 0.09381968528032303
2352
+ },
2353
+ {
2354
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2355
+ "layer": 17,
2356
+ "column": 2939,
2357
+ "static_cosine": 0.09380749613046646
2358
+ },
2359
+ {
2360
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
2361
+ "layer": 20,
2362
+ "column": 2191,
2363
+ "static_cosine": 0.09371107071638107
2364
+ },
2365
+ {
2366
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2367
+ "layer": 19,
2368
+ "column": 2842,
2369
+ "static_cosine": 0.0934593677520752
2370
+ },
2371
+ {
2372
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
2373
+ "layer": 15,
2374
+ "column": 329,
2375
+ "static_cosine": 0.09342271089553833
2376
+ },
2377
+ {
2378
+ "tensor": "model.language_model.layers.13.mlp.down_proj.weight",
2379
+ "layer": 13,
2380
+ "column": 3104,
2381
+ "static_cosine": 0.0933825671672821
2382
+ },
2383
+ {
2384
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
2385
+ "layer": 14,
2386
+ "column": 494,
2387
+ "static_cosine": 0.09312502294778824
2388
+ },
2389
+ {
2390
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
2391
+ "layer": 12,
2392
+ "column": 2123,
2393
+ "static_cosine": 0.0927717536687851
2394
+ },
2395
+ {
2396
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2397
+ "layer": 21,
2398
+ "column": 3155,
2399
+ "static_cosine": 0.09276270866394043
2400
+ },
2401
+ {
2402
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2403
+ "layer": 19,
2404
+ "column": 3334,
2405
+ "static_cosine": 0.09251472353935242
2406
+ },
2407
+ {
2408
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2409
+ "layer": 17,
2410
+ "column": 3009,
2411
+ "static_cosine": 0.09242354333400726
2412
+ },
2413
+ {
2414
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2415
+ "layer": 21,
2416
+ "column": 430,
2417
+ "static_cosine": 0.0924108475446701
2418
+ },
2419
+ {
2420
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2421
+ "layer": 16,
2422
+ "column": 695,
2423
+ "static_cosine": 0.09231404960155487
2424
+ },
2425
+ {
2426
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2427
+ "layer": 23,
2428
+ "column": 2544,
2429
+ "static_cosine": 0.0920613631606102
2430
+ },
2431
+ {
2432
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2433
+ "layer": 16,
2434
+ "column": 2652,
2435
+ "static_cosine": 0.09204891324043274
2436
+ },
2437
+ {
2438
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2439
+ "layer": 23,
2440
+ "column": 126,
2441
+ "static_cosine": 0.09196946024894714
2442
+ },
2443
+ {
2444
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
2445
+ "layer": 18,
2446
+ "column": 2971,
2447
+ "static_cosine": 0.09181059896945953
2448
+ },
2449
+ {
2450
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2451
+ "layer": 16,
2452
+ "column": 852,
2453
+ "static_cosine": 0.09176477044820786
2454
+ },
2455
+ {
2456
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
2457
+ "layer": 20,
2458
+ "column": 938,
2459
+ "static_cosine": 0.09157450497150421
2460
+ },
2461
+ {
2462
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2463
+ "layer": 17,
2464
+ "column": 3165,
2465
+ "static_cosine": 0.09145025908946991
2466
+ },
2467
+ {
2468
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
2469
+ "layer": 12,
2470
+ "column": 2178,
2471
+ "static_cosine": 0.09137703478336334
2472
+ },
2473
+ {
2474
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2475
+ "layer": 19,
2476
+ "column": 1133,
2477
+ "static_cosine": 0.09136662632226944
2478
+ },
2479
+ {
2480
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
2481
+ "layer": 15,
2482
+ "column": 597,
2483
+ "static_cosine": 0.09129209816455841
2484
+ },
2485
+ {
2486
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2487
+ "layer": 22,
2488
+ "column": 230,
2489
+ "static_cosine": 0.09103567898273468
2490
+ },
2491
+ {
2492
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2493
+ "layer": 23,
2494
+ "column": 3257,
2495
+ "static_cosine": 0.09088154137134552
2496
+ },
2497
+ {
2498
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2499
+ "layer": 23,
2500
+ "column": 3447,
2501
+ "static_cosine": 0.09066780656576157
2502
+ },
2503
+ {
2504
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
2505
+ "layer": 18,
2506
+ "column": 157,
2507
+ "static_cosine": 0.09060004353523254
2508
+ },
2509
+ {
2510
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2511
+ "layer": 23,
2512
+ "column": 60,
2513
+ "static_cosine": 0.09043717384338379
2514
+ },
2515
+ {
2516
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2517
+ "layer": 23,
2518
+ "column": 2523,
2519
+ "static_cosine": 0.09014892578125
2520
+ },
2521
+ {
2522
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2523
+ "layer": 16,
2524
+ "column": 2563,
2525
+ "static_cosine": 0.0900188758969307
2526
+ },
2527
+ {
2528
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2529
+ "layer": 21,
2530
+ "column": 2224,
2531
+ "static_cosine": 0.08992043882608414
2532
+ },
2533
+ {
2534
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2535
+ "layer": 23,
2536
+ "column": 454,
2537
+ "static_cosine": 0.0899178683757782
2538
+ },
2539
+ {
2540
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2541
+ "layer": 17,
2542
+ "column": 1494,
2543
+ "static_cosine": 0.08986648172140121
2544
+ },
2545
+ {
2546
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2547
+ "layer": 23,
2548
+ "column": 2367,
2549
+ "static_cosine": 0.08984152227640152
2550
+ },
2551
+ {
2552
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2553
+ "layer": 22,
2554
+ "column": 46,
2555
+ "static_cosine": 0.08972829580307007
2556
+ },
2557
+ {
2558
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2559
+ "layer": 22,
2560
+ "column": 3464,
2561
+ "static_cosine": 0.08971261978149414
2562
+ },
2563
+ {
2564
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2565
+ "layer": 16,
2566
+ "column": 2354,
2567
+ "static_cosine": 0.08948417752981186
2568
+ },
2569
+ {
2570
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2571
+ "layer": 21,
2572
+ "column": 821,
2573
+ "static_cosine": 0.08944113552570343
2574
+ },
2575
+ {
2576
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2577
+ "layer": 22,
2578
+ "column": 602,
2579
+ "static_cosine": 0.08944007009267807
2580
+ },
2581
+ {
2582
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
2583
+ "layer": 14,
2584
+ "column": 1016,
2585
+ "static_cosine": 0.08925766497850418
2586
+ },
2587
+ {
2588
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
2589
+ "layer": 15,
2590
+ "column": 83,
2591
+ "static_cosine": 0.08893260359764099
2592
+ },
2593
+ {
2594
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2595
+ "layer": 23,
2596
+ "column": 2234,
2597
+ "static_cosine": 0.08890016376972198
2598
+ },
2599
+ {
2600
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2601
+ "layer": 16,
2602
+ "column": 3532,
2603
+ "static_cosine": 0.08869770169258118
2604
+ },
2605
+ {
2606
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2607
+ "layer": 22,
2608
+ "column": 83,
2609
+ "static_cosine": 0.08864715695381165
2610
+ },
2611
+ {
2612
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2613
+ "layer": 22,
2614
+ "column": 3049,
2615
+ "static_cosine": 0.08845843374729156
2616
+ },
2617
+ {
2618
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2619
+ "layer": 17,
2620
+ "column": 2373,
2621
+ "static_cosine": 0.08843924105167389
2622
+ },
2623
+ {
2624
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2625
+ "layer": 16,
2626
+ "column": 2797,
2627
+ "static_cosine": 0.08825846016407013
2628
+ },
2629
+ {
2630
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2631
+ "layer": 16,
2632
+ "column": 1387,
2633
+ "static_cosine": 0.08820806443691254
2634
+ },
2635
+ {
2636
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
2637
+ "layer": 18,
2638
+ "column": 3372,
2639
+ "static_cosine": 0.08802701532840729
2640
+ },
2641
+ {
2642
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2643
+ "layer": 19,
2644
+ "column": 342,
2645
+ "static_cosine": 0.0879165455698967
2646
+ },
2647
+ {
2648
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
2649
+ "layer": 18,
2650
+ "column": 2999,
2651
+ "static_cosine": 0.08782291412353516
2652
+ },
2653
+ {
2654
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
2655
+ "layer": 12,
2656
+ "column": 2398,
2657
+ "static_cosine": 0.08772362768650055
2658
+ },
2659
+ {
2660
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2661
+ "layer": 19,
2662
+ "column": 461,
2663
+ "static_cosine": 0.08770786225795746
2664
+ },
2665
+ {
2666
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
2667
+ "layer": 18,
2668
+ "column": 1885,
2669
+ "static_cosine": 0.08769828081130981
2670
+ },
2671
+ {
2672
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2673
+ "layer": 22,
2674
+ "column": 3121,
2675
+ "static_cosine": 0.08754860609769821
2676
+ },
2677
+ {
2678
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
2679
+ "layer": 18,
2680
+ "column": 924,
2681
+ "static_cosine": 0.08741229772567749
2682
+ },
2683
+ {
2684
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2685
+ "layer": 23,
2686
+ "column": 19,
2687
+ "static_cosine": 0.08717925101518631
2688
+ },
2689
+ {
2690
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2691
+ "layer": 22,
2692
+ "column": 1533,
2693
+ "static_cosine": 0.08707685023546219
2694
+ },
2695
+ {
2696
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
2697
+ "layer": 18,
2698
+ "column": 3119,
2699
+ "static_cosine": 0.08700084686279297
2700
+ },
2701
+ {
2702
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
2703
+ "layer": 14,
2704
+ "column": 1281,
2705
+ "static_cosine": 0.08696988970041275
2706
+ },
2707
+ {
2708
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2709
+ "layer": 19,
2710
+ "column": 2509,
2711
+ "static_cosine": 0.0869571715593338
2712
+ },
2713
+ {
2714
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
2715
+ "layer": 20,
2716
+ "column": 2420,
2717
+ "static_cosine": 0.08691605180501938
2718
+ },
2719
+ {
2720
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
2721
+ "layer": 15,
2722
+ "column": 2166,
2723
+ "static_cosine": 0.0869000256061554
2724
+ },
2725
+ {
2726
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2727
+ "layer": 23,
2728
+ "column": 1578,
2729
+ "static_cosine": 0.08689926564693451
2730
+ },
2731
+ {
2732
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2733
+ "layer": 21,
2734
+ "column": 3360,
2735
+ "static_cosine": 0.08686449378728867
2736
+ },
2737
+ {
2738
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
2739
+ "layer": 14,
2740
+ "column": 3118,
2741
+ "static_cosine": 0.08678266406059265
2742
+ },
2743
+ {
2744
+ "tensor": "model.language_model.layers.13.mlp.down_proj.weight",
2745
+ "layer": 13,
2746
+ "column": 1419,
2747
+ "static_cosine": 0.08645259588956833
2748
+ },
2749
+ {
2750
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
2751
+ "layer": 20,
2752
+ "column": 524,
2753
+ "static_cosine": 0.08623870462179184
2754
+ },
2755
+ {
2756
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
2757
+ "layer": 14,
2758
+ "column": 2930,
2759
+ "static_cosine": 0.08604171872138977
2760
+ },
2761
+ {
2762
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2763
+ "layer": 23,
2764
+ "column": 1960,
2765
+ "static_cosine": 0.08603855222463608
2766
+ },
2767
+ {
2768
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2769
+ "layer": 23,
2770
+ "column": 1224,
2771
+ "static_cosine": 0.08602259308099747
2772
+ },
2773
+ {
2774
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
2775
+ "layer": 14,
2776
+ "column": 263,
2777
+ "static_cosine": 0.08597617596387863
2778
+ },
2779
+ {
2780
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
2781
+ "layer": 14,
2782
+ "column": 1474,
2783
+ "static_cosine": 0.08593782782554626
2784
+ },
2785
+ {
2786
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
2787
+ "layer": 18,
2788
+ "column": 3096,
2789
+ "static_cosine": 0.08584289997816086
2790
+ },
2791
+ {
2792
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
2793
+ "layer": 12,
2794
+ "column": 2712,
2795
+ "static_cosine": 0.08582289516925812
2796
+ },
2797
+ {
2798
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2799
+ "layer": 17,
2800
+ "column": 3146,
2801
+ "static_cosine": 0.08582240343093872
2802
+ },
2803
+ {
2804
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
2805
+ "layer": 20,
2806
+ "column": 2515,
2807
+ "static_cosine": 0.0858084037899971
2808
+ },
2809
+ {
2810
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
2811
+ "layer": 12,
2812
+ "column": 2335,
2813
+ "static_cosine": 0.0858018696308136
2814
+ },
2815
+ {
2816
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2817
+ "layer": 19,
2818
+ "column": 2339,
2819
+ "static_cosine": 0.08577001839876175
2820
+ },
2821
+ {
2822
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2823
+ "layer": 22,
2824
+ "column": 1113,
2825
+ "static_cosine": 0.08569622039794922
2826
+ },
2827
+ {
2828
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
2829
+ "layer": 20,
2830
+ "column": 3476,
2831
+ "static_cosine": 0.08548136055469513
2832
+ },
2833
+ {
2834
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2835
+ "layer": 22,
2836
+ "column": 6,
2837
+ "static_cosine": 0.08544189482927322
2838
+ },
2839
+ {
2840
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
2841
+ "layer": 20,
2842
+ "column": 3498,
2843
+ "static_cosine": 0.08535712957382202
2844
+ },
2845
+ {
2846
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2847
+ "layer": 17,
2848
+ "column": 165,
2849
+ "static_cosine": 0.08533669263124466
2850
+ },
2851
+ {
2852
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
2853
+ "layer": 18,
2854
+ "column": 385,
2855
+ "static_cosine": 0.08529245108366013
2856
+ },
2857
+ {
2858
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2859
+ "layer": 22,
2860
+ "column": 1495,
2861
+ "static_cosine": 0.08527716249227524
2862
+ },
2863
+ {
2864
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2865
+ "layer": 17,
2866
+ "column": 1410,
2867
+ "static_cosine": 0.08523781597614288
2868
+ },
2869
+ {
2870
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
2871
+ "layer": 15,
2872
+ "column": 847,
2873
+ "static_cosine": 0.08499214053153992
2874
+ },
2875
+ {
2876
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2877
+ "layer": 23,
2878
+ "column": 1198,
2879
+ "static_cosine": 0.08490894734859467
2880
+ },
2881
+ {
2882
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2883
+ "layer": 17,
2884
+ "column": 3394,
2885
+ "static_cosine": 0.08487461507320404
2886
+ },
2887
+ {
2888
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
2889
+ "layer": 19,
2890
+ "column": 620,
2891
+ "static_cosine": 0.08482439070940018
2892
+ },
2893
+ {
2894
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
2895
+ "layer": 15,
2896
+ "column": 2032,
2897
+ "static_cosine": 0.0847989022731781
2898
+ },
2899
+ {
2900
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
2901
+ "layer": 14,
2902
+ "column": 2998,
2903
+ "static_cosine": 0.08476439118385315
2904
+ },
2905
+ {
2906
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2907
+ "layer": 23,
2908
+ "column": 1826,
2909
+ "static_cosine": 0.08471766114234924
2910
+ },
2911
+ {
2912
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2913
+ "layer": 17,
2914
+ "column": 1341,
2915
+ "static_cosine": 0.08459744602441788
2916
+ },
2917
+ {
2918
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2919
+ "layer": 21,
2920
+ "column": 2001,
2921
+ "static_cosine": 0.08458410948514938
2922
+ },
2923
+ {
2924
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2925
+ "layer": 23,
2926
+ "column": 2317,
2927
+ "static_cosine": 0.08453361690044403
2928
+ },
2929
+ {
2930
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
2931
+ "layer": 17,
2932
+ "column": 974,
2933
+ "static_cosine": 0.08435258269309998
2934
+ },
2935
+ {
2936
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
2937
+ "layer": 20,
2938
+ "column": 2266,
2939
+ "static_cosine": 0.08417630195617676
2940
+ },
2941
+ {
2942
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
2943
+ "layer": 16,
2944
+ "column": 2162,
2945
+ "static_cosine": 0.08402256667613983
2946
+ },
2947
+ {
2948
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2949
+ "layer": 21,
2950
+ "column": 3537,
2951
+ "static_cosine": 0.0836852416396141
2952
+ },
2953
+ {
2954
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
2955
+ "layer": 23,
2956
+ "column": 3352,
2957
+ "static_cosine": 0.08362327516078949
2958
+ },
2959
+ {
2960
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
2961
+ "layer": 15,
2962
+ "column": 39,
2963
+ "static_cosine": 0.08360247313976288
2964
+ },
2965
+ {
2966
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
2967
+ "layer": 15,
2968
+ "column": 2128,
2969
+ "static_cosine": 0.08355815708637238
2970
+ },
2971
+ {
2972
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
2973
+ "layer": 22,
2974
+ "column": 19,
2975
+ "static_cosine": 0.08353157341480255
2976
+ },
2977
+ {
2978
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
2979
+ "layer": 12,
2980
+ "column": 1000,
2981
+ "static_cosine": 0.08349194377660751
2982
+ },
2983
+ {
2984
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2985
+ "layer": 21,
2986
+ "column": 3532,
2987
+ "static_cosine": 0.08330792933702469
2988
+ },
2989
+ {
2990
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
2991
+ "layer": 21,
2992
+ "column": 1390,
2993
+ "static_cosine": 0.08322285115718842
2994
+ },
2995
+ {
2996
+ "tensor": "model.language_model.layers.13.mlp.down_proj.weight",
2997
+ "layer": 13,
2998
+ "column": 3566,
2999
+ "static_cosine": 0.08317442238330841
3000
+ },
3001
+ {
3002
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
3003
+ "layer": 21,
3004
+ "column": 2430,
3005
+ "static_cosine": 0.08314302563667297
3006
+ },
3007
+ {
3008
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
3009
+ "layer": 15,
3010
+ "column": 446,
3011
+ "static_cosine": 0.08307268470525742
3012
+ },
3013
+ {
3014
+ "tensor": "model.language_model.layers.13.mlp.down_proj.weight",
3015
+ "layer": 13,
3016
+ "column": 952,
3017
+ "static_cosine": 0.08297602832317352
3018
+ },
3019
+ {
3020
+ "tensor": "model.language_model.layers.15.mlp.down_proj.weight",
3021
+ "layer": 15,
3022
+ "column": 194,
3023
+ "static_cosine": 0.08295321464538574
3024
+ },
3025
+ {
3026
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
3027
+ "layer": 16,
3028
+ "column": 1825,
3029
+ "static_cosine": 0.08285076916217804
3030
+ },
3031
+ {
3032
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
3033
+ "layer": 20,
3034
+ "column": 600,
3035
+ "static_cosine": 0.08270015567541122
3036
+ },
3037
+ {
3038
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
3039
+ "layer": 14,
3040
+ "column": 2015,
3041
+ "static_cosine": 0.0825992226600647
3042
+ },
3043
+ {
3044
+ "tensor": "model.language_model.layers.13.mlp.down_proj.weight",
3045
+ "layer": 13,
3046
+ "column": 1198,
3047
+ "static_cosine": 0.08250514417886734
3048
+ },
3049
+ {
3050
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
3051
+ "layer": 22,
3052
+ "column": 1617,
3053
+ "static_cosine": 0.08249792456626892
3054
+ },
3055
+ {
3056
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
3057
+ "layer": 18,
3058
+ "column": 338,
3059
+ "static_cosine": 0.08248809725046158
3060
+ },
3061
+ {
3062
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
3063
+ "layer": 22,
3064
+ "column": 800,
3065
+ "static_cosine": 0.0824853926897049
3066
+ },
3067
+ {
3068
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
3069
+ "layer": 12,
3070
+ "column": 531,
3071
+ "static_cosine": 0.0824466347694397
3072
+ },
3073
+ {
3074
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
3075
+ "layer": 21,
3076
+ "column": 1622,
3077
+ "static_cosine": 0.08241860568523407
3078
+ },
3079
+ {
3080
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
3081
+ "layer": 20,
3082
+ "column": 1240,
3083
+ "static_cosine": 0.08234641700983047
3084
+ },
3085
+ {
3086
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
3087
+ "layer": 14,
3088
+ "column": 1622,
3089
+ "static_cosine": 0.08228050172328949
3090
+ },
3091
+ {
3092
+ "tensor": "model.language_model.layers.20.mlp.down_proj.weight",
3093
+ "layer": 20,
3094
+ "column": 2088,
3095
+ "static_cosine": 0.08223514258861542
3096
+ },
3097
+ {
3098
+ "tensor": "model.language_model.layers.12.mlp.down_proj.weight",
3099
+ "layer": 12,
3100
+ "column": 2458,
3101
+ "static_cosine": 0.08218204230070114
3102
+ },
3103
+ {
3104
+ "tensor": "model.language_model.layers.13.mlp.down_proj.weight",
3105
+ "layer": 13,
3106
+ "column": 992,
3107
+ "static_cosine": 0.08215226233005524
3108
+ },
3109
+ {
3110
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
3111
+ "layer": 22,
3112
+ "column": 2241,
3113
+ "static_cosine": 0.08210191875696182
3114
+ },
3115
+ {
3116
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
3117
+ "layer": 17,
3118
+ "column": 3073,
3119
+ "static_cosine": 0.08208631724119186
3120
+ },
3121
+ {
3122
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
3123
+ "layer": 22,
3124
+ "column": 2828,
3125
+ "static_cosine": 0.08196385949850082
3126
+ },
3127
+ {
3128
+ "tensor": "model.language_model.layers.23.mlp.down_proj.weight",
3129
+ "layer": 23,
3130
+ "column": 2720,
3131
+ "static_cosine": 0.08193042874336243
3132
+ },
3133
+ {
3134
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
3135
+ "layer": 17,
3136
+ "column": 3349,
3137
+ "static_cosine": 0.08190808445215225
3138
+ },
3139
+ {
3140
+ "tensor": "model.language_model.layers.21.mlp.down_proj.weight",
3141
+ "layer": 21,
3142
+ "column": 2323,
3143
+ "static_cosine": 0.08184511959552765
3144
+ },
3145
+ {
3146
+ "tensor": "model.language_model.layers.16.mlp.down_proj.weight",
3147
+ "layer": 16,
3148
+ "column": 3526,
3149
+ "static_cosine": 0.08180224150419235
3150
+ },
3151
+ {
3152
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
3153
+ "layer": 22,
3154
+ "column": 1415,
3155
+ "static_cosine": 0.08175577968358994
3156
+ },
3157
+ {
3158
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
3159
+ "layer": 14,
3160
+ "column": 1694,
3161
+ "static_cosine": 0.08173462748527527
3162
+ },
3163
+ {
3164
+ "tensor": "model.language_model.layers.19.mlp.down_proj.weight",
3165
+ "layer": 19,
3166
+ "column": 3250,
3167
+ "static_cosine": 0.08169515430927277
3168
+ },
3169
+ {
3170
+ "tensor": "model.language_model.layers.22.mlp.down_proj.weight",
3171
+ "layer": 22,
3172
+ "column": 2524,
3173
+ "static_cosine": 0.08164513856172562
3174
+ },
3175
+ {
3176
+ "tensor": "model.language_model.layers.18.mlp.down_proj.weight",
3177
+ "layer": 18,
3178
+ "column": 3208,
3179
+ "static_cosine": 0.08162808418273926
3180
+ },
3181
+ {
3182
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
3183
+ "layer": 14,
3184
+ "column": 2216,
3185
+ "static_cosine": 0.0816117376089096
3186
+ },
3187
+ {
3188
+ "tensor": "model.language_model.layers.14.mlp.down_proj.weight",
3189
+ "layer": 14,
3190
+ "column": 3544,
3191
+ "static_cosine": 0.08151750266551971
3192
+ },
3193
+ {
3194
+ "tensor": "model.language_model.layers.17.mlp.down_proj.weight",
3195
+ "layer": 17,
3196
+ "column": 734,
3197
+ "static_cosine": 0.08148801326751709
3198
+ }
3199
+ ],
3200
+ "verification": {
3201
+ "all_tensors_checked": 488,
3202
+ "unchanged_tensors": 476,
3203
+ "changed_tensors": 12,
3204
+ "changed_scalars": 65536,
3205
+ "total_stored_parameters": 873438784,
3206
+ "changed_fraction": 7.503216161282804e-05,
3207
+ "unselected_bytes_identical": true,
3208
+ "original_checkpoint_unchanged": true,
3209
+ "config_and_tokenizer_files_identical": true,
3210
+ "original_sha256": "04b1c301231dd422b8860db31311ab2721511346a32cb1e079c4c4e5f1fe4696",
3211
+ "edited_sha256": "12d6e276309afe792154864b2f96dd23ed6b75d8f403d6dd9029a5388901e123"
3212
+ },
3213
+ "sources": [
3214
+ "https://huggingface.co/Qwen/Qwen3.5-0.8B",
3215
+ "https://aclanthology.org/2021.emnlp-main.446/"
3216
+ ],
3217
+ "source_scope": "Geva et al. motivates static inspection of FFN value vectors, not this edit or its effectiveness on Qwen3.5/Blocksworld."
3218
+ }
video_preprocessor_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "size": {
3
+ "longest_edge": 25165824,
4
+ "shortest_edge": 4096
5
+ },
6
+ "patch_size": 16,
7
+ "temporal_patch_size": 2,
8
+ "merge_size": 2,
9
+ "image_mean": [
10
+ 0.5,
11
+ 0.5,
12
+ 0.5
13
+ ],
14
+ "image_std": [
15
+ 0.5,
16
+ 0.5,
17
+ 0.5
18
+ ],
19
+ "processor_class": "Qwen3VLProcessor",
20
+ "video_processor_type": "Qwen3VLVideoProcessor"
21
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff