Hugging Face's logo

sassoftware
/
glitext-medium

Token Classification
PyTorch
ONNX
multilingual
glitext
Model card Files
xet
 Community
barry-sas commited on
Commit
fd904d9
·
1 Parent(s): 2cafc97

Add onnx files

Browse files
Files changed (7) hide show
  1. LICENSE +202 -0
  2. README.md +32 -74
  3. gliner_config.json +61 -3
  4. model.onnx +3 -0
  5. modelaudit.json +1073 -0
  6. tokenizer.json +0 -0
  7. tokenizer_config.json +4 -58
LICENSE ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ Apache License
3
+ Version 2.0, January 2004
4
+ http://www.apache.org/licenses/
5
+
6
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
7
+
8
+ 1. Definitions.
9
+
10
+ "License" shall mean the terms and conditions for use, reproduction,
11
+ and distribution as defined by Sections 1 through 9 of this document.
12
+
13
+ "Licensor" shall mean the copyright owner or entity authorized by
14
+ the copyright owner that is granting the License.
15
+
16
+ "Legal Entity" shall mean the union of the acting entity and all
17
+ other entities that control, are controlled by, or are under common
18
+ control with that entity. For the purposes of this definition,
19
+ "control" means (i) the power, direct or indirect, to cause the
20
+ direction or management of such entity, whether by contract or
21
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
22
+ outstanding shares, or (iii) beneficial ownership of such entity.
23
+
24
+ "You" (or "Your") shall mean an individual or Legal Entity
25
+ exercising permissions granted by this License.
26
+
27
+ "Source" form shall mean the preferred form for making modifications,
28
+ including but not limited to software source code, documentation
29
+ source, and configuration files.
30
+
31
+ "Object" form shall mean any form resulting from mechanical
32
+ transformation or translation of a Source form, including but
33
+ not limited to compiled object code, generated documentation,
34
+ and conversions to other media types.
35
+
36
+ "Work" shall mean the work of authorship, whether in Source or
37
+ Object form, made available under the License, as indicated by a
38
+ copyright notice that is included in or attached to the work
39
+ (an example is provided in the Appendix below).
40
+
41
+ "Derivative Works" shall mean any work, whether in Source or Object
42
+ form, that is based on (or derived from) the Work and for which the
43
+ editorial revisions, annotations, elaborations, or other modifications
44
+ represent, as a whole, an original work of authorship. For the purposes
45
+ of this License, Derivative Works shall not include works that remain
46
+ separable from, or merely link (or bind by name) to the interfaces of,
47
+ the Work and Derivative Works thereof.
48
+
49
+ "Contribution" shall mean any work of authorship, including
50
+ the original version of the Work and any modifications or additions
51
+ to that Work or Derivative Works thereof, that is intentionally
52
+ submitted to Licensor for inclusion in the Work by the copyright owner
53
+ or by an individual or Legal Entity authorized to submit on behalf of
54
+ the copyright owner. For the purposes of this definition, "submitted"
55
+ means any form of electronic, verbal, or written communication sent
56
+ to the Licensor or its representatives, including but not limited to
57
+ communication on electronic mailing lists, source code control systems,
58
+ and issue tracking systems that are managed by, or on behalf of, the
59
+ Licensor for the purpose of discussing and improving the Work, but
60
+ excluding communication that is conspicuously marked or otherwise
61
+ designated in writing by the copyright owner as "Not a Contribution."
62
+
63
+ "Contributor" shall mean Licensor and any individual or Legal Entity
64
+ on behalf of whom a Contribution has been received by Licensor and
65
+ subsequently incorporated within the Work.
66
+
67
+ 2. Grant of Copyright License. Subject to the terms and conditions of
68
+ this License, each Contributor hereby grants to You a perpetual,
69
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
70
+ copyright license to reproduce, prepare Derivative Works of,
71
+ publicly display, publicly perform, sublicense, and distribute the
72
+ Work and such Derivative Works in Source or Object form.
73
+
74
+ 3. Grant of Patent License. Subject to the terms and conditions of
75
+ this License, each Contributor hereby grants to You a perpetual,
76
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
77
+ (except as stated in this section) patent license to make, have made,
78
+ use, offer to sell, sell, import, and otherwise transfer the Work,
79
+ where such license applies only to those patent claims licensable
80
+ by such Contributor that are necessarily infringed by their
81
+ Contribution(s) alone or by combination of their Contribution(s)
82
+ with the Work to which such Contribution(s) was submitted. If You
83
+ institute patent litigation against any entity (including a
84
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
85
+ or a Contribution incorporated within the Work constitutes direct
86
+ or contributory patent infringement, then any patent licenses
87
+ granted to You under this License for that Work shall terminate
88
+ as of the date such litigation is filed.
89
+
90
+ 4. Redistribution. You may reproduce and distribute copies of the
91
+ Work or Derivative Works thereof in any medium, with or without
92
+ modifications, and in Source or Object form, provided that You
93
+ meet the following conditions:
94
+
95
+ (a) You must give any other recipients of the Work or
96
+ Derivative Works a copy of this License; and
97
+
98
+ (b) You must cause any modified files to carry prominent notices
99
+ stating that You changed the files; and
100
+
101
+ (c) You must retain, in the Source form of any Derivative Works
102
+ that You distribute, all copyright, patent, trademark, and
103
+ attribution notices from the Source form of the Work,
104
+ excluding those notices that do not pertain to any part of
105
+ the Derivative Works; and
106
+
107
+ (d) If the Work includes a "NOTICE" text file as part of its
108
+ distribution, then any Derivative Works that You distribute must
109
+ include a readable copy of the attribution notices contained
110
+ within such NOTICE file, excluding those notices that do not
111
+ pertain to any part of the Derivative Works, in at least one
112
+ of the following places: within a NOTICE text file distributed
113
+ as part of the Derivative Works; within the Source form or
114
+ documentation, if provided along with the Derivative Works; or,
115
+ within a display generated by the Derivative Works, if and
116
+ wherever such third-party notices normally appear. The contents
117
+ of the NOTICE file are for informational purposes only and
118
+ do not modify the License. You may add Your own attribution
119
+ notices within Derivative Works that You distribute, alongside
120
+ or as an addendum to the NOTICE text from the Work, provided
121
+ that such additional attribution notices cannot be construed
122
+ as modifying the License.
123
+
124
+ You may add Your own copyright statement to Your modifications and
125
+ may provide additional or different license terms and conditions
126
+ for use, reproduction, or distribution of Your modifications, or
127
+ for any such Derivative Works as a whole, provided Your use,
128
+ reproduction, and distribution of the Work otherwise complies with
129
+ the conditions stated in this License.
130
+
131
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
132
+ any Contribution intentionally submitted for inclusion in the Work
133
+ by You to the Licensor shall be under the terms and conditions of
134
+ this License, without any additional terms or conditions.
135
+ Notwithstanding the above, nothing herein shall supersede or modify
136
+ the terms of any separate license agreement you may have executed
137
+ with Licensor regarding such Contributions.
138
+
139
+ 6. Trademarks. This License does not grant permission to use the trade
140
+ names, trademarks, service marks, or product names of the Licensor,
141
+ except as required for reasonable and customary use in describing the
142
+ origin of the Work and reproducing the content of the NOTICE file.
143
+
144
+ 7. Disclaimer of Warranty. Unless required by applicable law or
145
+ agreed to in writing, Licensor provides the Work (and each
146
+ Contributor provides its Contributions) on an "AS IS" BASIS,
147
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
148
+ implied, including, without limitation, any warranties or conditions
149
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
150
+ PARTICULAR PURPOSE. You are solely responsible for determining the
151
+ appropriateness of using or redistributing the Work and assume any
152
+ risks associated with Your exercise of permissions under this License.
153
+
154
+ 8. Limitation of Liability. In no event and under no legal theory,
155
+ whether in tort (including negligence), contract, or otherwise,
156
+ unless required by applicable law (such as deliberate and grossly
157
+ negligent acts) or agreed to in writing, shall any Contributor be
158
+ liable to You for damages, including any direct, indirect, special,
159
+ incidental, or consequential damages of any character arising as a
160
+ result of this License or out of the use or inability to use the
161
+ Work (including but not limited to damages for loss of goodwill,
162
+ work stoppage, computer failure or malfunction, or any and all
163
+ other commercial damages or losses), even if such Contributor
164
+ has been advised of the possibility of such damages.
165
+
166
+ 9. Accepting Warranty or Additional Liability. While redistributing
167
+ the Work or Derivative Works thereof, You may choose to offer,
168
+ and charge a fee for, acceptance of support, warranty, indemnity,
169
+ or other liability obligations and/or rights consistent with this
170
+ License. However, in accepting such obligations, You may act only
171
+ on Your own behalf and on Your sole responsibility, not on behalf
172
+ of any other Contributor, and only if You agree to indemnify,
173
+ defend, and hold each Contributor harmless for any liability
174
+ incurred by, or claims asserted against, such Contributor by reason
175
+ of your accepting any such warranty or additional liability.
176
+
177
+ END OF TERMS AND CONDITIONS
178
+
179
+ APPENDIX: How to apply the Apache License to your work.
180
+
181
+ To apply the Apache License to your work, attach the following
182
+ boilerplate notice, with the fields enclosed by brackets "[]"
183
+ replaced with your own identifying information. (Don't include
184
+ the brackets!) The text should be enclosed in the appropriate
185
+ comment syntax for the file format. We also recommend that a
186
+ file or class name and description of purpose be included on the
187
+ same "printed page" as the copyright notice for easier
188
+ identification within third-party archives.
189
+
190
+ Copyright [yyyy] [name of copyright owner]
191
+
192
+ Licensed under the Apache License, Version 2.0 (the "License");
193
+ you may not use this file except in compliance with the License.
194
+ You may obtain a copy of the License at
195
+
196
+ http://www.apache.org/licenses/LICENSE-2.0
197
+
198
+ Unless required by applicable law or agreed to in writing, software
199
+ distributed under the License is distributed on an "AS IS" BASIS,
200
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
201
+ See the License for the specific language governing permissions and
202
+ limitations under the License.
README.md CHANGED
@@ -1,95 +1,53 @@
1
  ---
 
2
  license: apache-2.0
3
- language:
4
- - multilingual
5
- library_name: gliner
6
- datasets:
7
- - urchade/pile-mistral-v0.1
8
- pipeline_tag: token-classification
 
 
 
 
 
 
 
9
  ---
10
 
11
- # About
12
 
13
- GLiNER is a Named Entity Recognition (NER) model capable of identifying any entity type using a bidirectional transformer encoder (BERT-like). It provides a practical alternative to traditional NER models, which are limited to predefined entities, and Large Language Models (LLMs) that, despite their flexibility, are costly and large for resource-constrained scenarios.
14
 
 
15
 
16
- ## Links
17
 
18
- * Paper: https://arxiv.org/abs/2311.08526
19
- * Repository: https://github.com/urchade/GLiNER
20
-
21
- ## Installation
22
- To use this model, you must install the GLiNER Python library:
23
  ```
24
- !pip install gliner -U
25
  ```
26
 
27
- ## Usage
28
- Once you've downloaded the GLiNER library, you can import the GLiNER class. You can then load this model using `GLiNER.from_pretrained` and predict entities with `predict_entities`.
29
-
30
- ```python
31
- from gliner import GLiNER
32
-
33
- model = GLiNER.from_pretrained("gliner-community/gliner_medium-v2.5", load_tokenizer=True)
34
-
35
- text = """
36
- Cristiano Ronaldo dos Santos Aveiro (Portuguese pronunciation: [kɾiʃˈtjɐnu ʁɔˈnaldu]; born 5 February 1985) is a Portuguese professional footballer who plays as a forward for and captains both Saudi Pro League club Al Nassr and the Portugal national team. Widely regarded as one of the greatest players of all time, Ronaldo has won five Ballon d'Or awards,[note 3] a record three UEFA Men's Player of the Year Awards, and four European Golden Shoes, the most by a European player. He has won 33 trophies in his career, including seven league titles, five UEFA Champions Leagues, the UEFA European Championship and the UEFA Nations League. Ronaldo holds the records for most appearances (183), goals (140) and assists (42) in the Champions League, goals in the European Championship (14), international goals (128) and international appearances (205). He is one of the few players to have made over 1,200 professional career appearances, the most by an outfield player, and has scored over 850 official senior career goals for club and country, making him the top goalscorer of all time.
37
- """
38
-
39
- labels = ["person", "award", "date", "competitions", "teams"]
40
-
41
- entities = model.predict_entities(text, labels)
42
-
43
- for entity in entities:
44
- print(entity["text"], "=>", entity["label"])
45
- ```
46
 
47
  ```
48
- Cristiano Ronaldo dos Santos Aveiro => person
49
- 5 February 1985 => date
50
- Al Nassr => teams
51
- Portugal national team => teams
52
- Ballon d'Or => award
53
- UEFA Men's Player of the Year Awards => award
54
- European Golden Shoes => award
55
- UEFA Champions Leagues => competitions
56
- UEFA European Championship => competitions
57
- UEFA Nations League => competitions
58
- Champions League => competitions
59
- European Championship => competitions
60
  ```
61
 
62
- ## Named Entity Recognition benchmark result
63
- Below is a comparison of results between previous versions of the model and the current one:
64
- ![Models performance](models_comparison.png)
 
65
 
 
66
 
67
- ## Available models
68
 
69
- | Release | Model Name | # of Parameters | Language | License |
70
- | - | - | - | - | - |
71
- | v0 | [urchade/gliner_base](https://huggingface.co/urchade/gliner_base)<br>[urchade/gliner_multi](https://huggingface.co/urchade/gliner_multi) | 209M<br>209M | English<br>Multilingual | cc-by-nc-4.0 |
72
- | v1 | [urchade/gliner_small-v1](https://huggingface.co/urchade/gliner_small-v1)<br>[urchade/gliner_medium-v1](https://huggingface.co/urchade/gliner_medium-v1)<br>[urchade/gliner_large-v1](https://huggingface.co/urchade/gliner_large-v1) | 166M<br>209M<br>459M | English <br> English <br> English | cc-by-nc-4.0 |
73
- | v2 | [urchade/gliner_small-v2](https://huggingface.co/urchade/gliner_small-v2)<br>[urchade/gliner_medium-v2](https://huggingface.co/urchade/gliner_medium-v2)<br>[urchade/gliner_large-v2](https://huggingface.co/urchade/gliner_large-v2) | 166M<br>209M<br>459M | English <br> English <br> English | apache-2.0 |
74
- | v2.1 | [urchade/gliner_small-v2.1](https://huggingface.co/urchade/gliner_small-v2.1)<br>[urchade/gliner_medium-v2.1](https://huggingface.co/urchade/gliner_medium-v2.1)<br>[urchade/gliner_large-v2.1](https://huggingface.co/urchade/gliner_large-v2.1) <br>[urchade/gliner_multi-v2.1](https://huggingface.co/urchade/gliner_multi-v2.1) | 166M<br>209M<br>459M<br>209M | English <br> English <br> English <br> Multilingual | apache-2.0 |
75
 
 
 
 
76
 
77
- ## Model Authors
78
- The model authors are:
79
- * [Urchade Zaratiana](https://huggingface.co/urchade)
80
- * [Ihor Stepanov](https://huggingface.co/Ihor)
81
- * Nadi Tomeh
82
- * Pierre Holat
83
- * Thierry Charnois
84
 
85
- ## Citation
86
- ```bibtex
87
- @misc{zaratiana2023gliner,
88
- title={GLiNER: Generalist Model for Named Entity Recognition using Bidirectional Transformer},
89
- author={Urchade Zaratiana and Nadi Tomeh and Pierre Holat and Thierry Charnois},
90
- year={2023},
91
- eprint={2311.08526},
92
- archivePrefix={arXiv},
93
- primaryClass={cs.CL}
94
- }
95
- ```
 
1
  ---
2
+ library_name: glitext
3
  license: apache-2.0
4
+ tags:
5
+ - glitext
6
+ glitext:
7
+ name: medium
8
+ label: GliText Recognition (Balanced)
9
+ description: An efficient zero-shot named entity recognition model tuned for generalized extraction with balanced speed and accuracy.
10
+ recognition: true
11
+ classification: false
12
+ association: false
13
+ span_mode: true
14
+ size_gb: 0.78
15
+ hf_repo: rpeel/glitext-medium
16
+ source_url: gliner-community/gliner_medium-v2.5
17
  ---
18
 
19
+ # rpeel/glitext-medium
20
 
21
+ An efficient zero-shot named entity recognition model tuned for generalized extraction with balanced speed and accuracy.
22
 
23
+ ## Requirements
24
 
25
+ To download this model to the SAS GLiText server:
26
 
 
 
 
 
 
27
  ```
28
+ POST /v1/models/download?name=medium
29
  ```
30
 
31
+ To download and load into memory in one step:
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
32
 
33
  ```
34
+ PUT /v1/models?name=medium
 
 
 
 
 
 
 
 
 
 
 
35
  ```
36
 
37
+ ## Source Model
38
+
39
+ Exported from [gliner-community/gliner_medium-v2.5](https://huggingface.co/gliner-community/gliner_medium-v2.5).
40
+ See the [original model card](https://huggingface.co/gliner-community/gliner_medium-v2.5) for full architecture and training details.
41
 
42
+ ## Security Scan
43
 
44
+ Scanned with [modelaudit](https://github.com/promptfoo/modelaudit) v0.2.40 on 2026-04-27. 24/24 checks passed. [Full results](modelaudit.json).
45
 
 
 
 
 
 
 
46
 
47
+ | File | Size | SHA-256 |
48
+ |------|------|---------|
49
+ | `model.onnx` | 835.6 MB | `dfbf82b4c9b7cb8e…` |
50
 
51
+ ## License
 
 
 
 
 
 
52
 
53
+ [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0). Derived from [gliner-community/gliner_medium-v2.5](https://huggingface.co/gliner-community/gliner_medium-v2.5) by [gliner-community](https://huggingface.co/gliner-community).
 
 
 
 
 
 
 
 
 
 
gliner_config.json CHANGED
@@ -1,9 +1,60 @@
1
  {
2
  "class_token_index": 128001,
3
  "dropout": 0.4,
4
- "encoder_config": null,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5
  "ent_token": "<<ENT>>",
6
  "fine_tune": true,
 
7
  "has_rnn": true,
8
  "hidden_size": 768,
9
  "label_smoothing": 0.0,
@@ -15,14 +66,21 @@
15
  "max_types": 30,
16
  "max_width": 12,
17
  "model_name": "microsoft/deberta-v3-base",
18
- "model_type": "gliner",
19
  "name": "span level gliner",
 
 
 
 
20
  "random_drop": true,
 
21
  "sep_token": "<<SEP>>",
22
  "shuffle_types": true,
 
23
  "span_mode": "markerV0",
24
  "subtoken_pooling": "first",
25
- "transformers_version": "4.40.2",
 
26
  "vocab_size": 128003,
27
  "words_splitter_type": "whitespace"
28
  }
 
1
  {
2
  "class_token_index": 128001,
3
  "dropout": 0.4,
4
+ "embed_ent_token": true,
5
+ "encoder_config": {
6
+ "_name_or_path": "microsoft/deberta-v3-base",
7
+ "architectures": null,
8
+ "attention_probs_dropout_prob": 0.1,
9
+ "bos_token_id": null,
10
+ "chunk_size_feed_forward": 0,
11
+ "dtype": null,
12
+ "eos_token_id": null,
13
+ "hidden_act": "gelu",
14
+ "hidden_dropout_prob": 0.1,
15
+ "hidden_size": 768,
16
+ "id2label": {
17
+ "0": "LABEL_0",
18
+ "1": "LABEL_1"
19
+ },
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 3072,
22
+ "is_encoder_decoder": false,
23
+ "label2id": {
24
+ "LABEL_0": 0,
25
+ "LABEL_1": 1
26
+ },
27
+ "layer_norm_eps": 1e-07,
28
+ "legacy": true,
29
+ "max_position_embeddings": 512,
30
+ "max_relative_positions": -1,
31
+ "model_type": "deberta-v2",
32
+ "norm_rel_ebd": "layer_norm",
33
+ "num_attention_heads": 12,
34
+ "num_hidden_layers": 12,
35
+ "output_attentions": false,
36
+ "output_hidden_states": false,
37
+ "pad_token_id": 0,
38
+ "pooler_dropout": 0,
39
+ "pooler_hidden_act": "gelu",
40
+ "pooler_hidden_size": 768,
41
+ "pos_att_type": [
42
+ "p2c",
43
+ "c2p"
44
+ ],
45
+ "position_biased_input": false,
46
+ "position_buckets": 256,
47
+ "problem_type": null,
48
+ "relative_attention": true,
49
+ "return_dict": true,
50
+ "share_att_key": true,
51
+ "tie_word_embeddings": true,
52
+ "type_vocab_size": 0,
53
+ "vocab_size": 128003
54
+ },
55
  "ent_token": "<<ENT>>",
56
  "fine_tune": true,
57
+ "fuse_layers": false,
58
  "has_rnn": true,
59
  "hidden_size": 768,
60
  "label_smoothing": 0.0,
 
66
  "max_types": 30,
67
  "max_width": 12,
68
  "model_name": "microsoft/deberta-v3-base",
69
+ "model_type": null,
70
  "name": "span level gliner",
71
+ "neg_spans_ratio": 1.0,
72
+ "num_post_fusion_layers": 1,
73
+ "num_rnn_layers": 1,
74
+ "post_fusion_schema": "",
75
  "random_drop": true,
76
+ "represent_spans": false,
77
  "sep_token": "<<SEP>>",
78
  "shuffle_types": true,
79
+ "span_loss_coef": 1.0,
80
  "span_mode": "markerV0",
81
  "subtoken_pooling": "first",
82
+ "token_loss_coef": 1.0,
83
+ "transformers_version": "5.1.0",
84
  "vocab_size": 128003,
85
  "words_splitter_type": "whitespace"
86
  }
model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfbf82b4c9b7cb8ebedb6e04bec068db7cfee3053cb90c7d4aa11b8a93edb46a
3
+ size 835590772
modelaudit.json ADDED
@@ -0,0 +1,1073 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "tool": "modelaudit",
3
+ "tool_version": "0.2.40",
4
+ "scanned_at": "2026-04-27T00:38:25Z",
5
+ "files": {
6
+ "model.onnx": {
7
+ "size_mb": 835.6,
8
+ "sha256": "dfbf82b4c9b7cb8ebedb6e04bec068db7cfee3053cb90c7d4aa11b8a93edb46a"
9
+ }
10
+ },
11
+ "audit": {
12
+ "bytes_scanned": 843939140,
13
+ "issues": [
14
+ {
15
+ "message": "Weight distribution analysis skipped one or more eligible ONNX initializers",
16
+ "severity": "info",
17
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
18
+ "details": {
19
+ "scan_outcome_reason": "onnx_weight_distribution_analysis_incomplete",
20
+ "coverage_gap": "partial_initializer_coverage",
21
+ "eligible_initializers": 85,
22
+ "analyzed_initializers": 84,
23
+ "external_initializers_skipped": 0,
24
+ "oversized_initializers_skipped": 1,
25
+ "extraction_failures": 0,
26
+ "max_array_size": 104857600
27
+ },
28
+ "timestamp": 1777250300.2642596,
29
+ "type": "onnx_check",
30
+ "rule_code": "S902"
31
+ },
32
+ {
33
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 7 has unusually dissimilar weights",
34
+ "severity": "info",
35
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
36
+ "details": {
37
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
38
+ "neuron_index": 7,
39
+ "max_similarity_to_others": 0.6353273391723633,
40
+ "weight_norm": 5.372602462768555,
41
+ "total_outputs": 768,
42
+ "analysis_method": "structural_analysis"
43
+ },
44
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
45
+ "timestamp": 1777250303.1872206,
46
+ "type": "onnx_check",
47
+ "rule_code": "S803"
48
+ },
49
+ {
50
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 48 has unusually dissimilar weights",
51
+ "severity": "info",
52
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
53
+ "details": {
54
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
55
+ "neuron_index": 48,
56
+ "max_similarity_to_others": 0.6696844100952148,
57
+ "weight_norm": 0.46152299642562866,
58
+ "total_outputs": 768,
59
+ "analysis_method": "structural_analysis"
60
+ },
61
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
62
+ "timestamp": 1777250303.1879716,
63
+ "type": "onnx_check",
64
+ "rule_code": "S803"
65
+ },
66
+ {
67
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 117 has unusually dissimilar weights",
68
+ "severity": "info",
69
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
70
+ "details": {
71
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
72
+ "neuron_index": 117,
73
+ "max_similarity_to_others": 0.634371280670166,
74
+ "weight_norm": 0.39997434616088867,
75
+ "total_outputs": 768,
76
+ "analysis_method": "structural_analysis"
77
+ },
78
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
79
+ "timestamp": 1777250303.1885126,
80
+ "type": "onnx_check",
81
+ "rule_code": "S803"
82
+ },
83
+ {
84
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 213 has unusually dissimilar weights",
85
+ "severity": "info",
86
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
87
+ "details": {
88
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
89
+ "neuron_index": 213,
90
+ "max_similarity_to_others": 0.6954333186149597,
91
+ "weight_norm": 0.5225619673728943,
92
+ "total_outputs": 768,
93
+ "analysis_method": "structural_analysis"
94
+ },
95
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
96
+ "timestamp": 1777250303.1890316,
97
+ "type": "onnx_check",
98
+ "rule_code": "S803"
99
+ },
100
+ {
101
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 248 has unusually dissimilar weights",
102
+ "severity": "info",
103
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
104
+ "details": {
105
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
106
+ "neuron_index": 248,
107
+ "max_similarity_to_others": 0.6133643388748169,
108
+ "weight_norm": 0.47258928418159485,
109
+ "total_outputs": 768,
110
+ "analysis_method": "structural_analysis"
111
+ },
112
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
113
+ "timestamp": 1777250303.1895173,
114
+ "type": "onnx_check",
115
+ "rule_code": "S803"
116
+ },
117
+ {
118
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 266 has unusually dissimilar weights",
119
+ "severity": "info",
120
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
121
+ "details": {
122
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
123
+ "neuron_index": 266,
124
+ "max_similarity_to_others": 0.649356484413147,
125
+ "weight_norm": 0.683588981628418,
126
+ "total_outputs": 768,
127
+ "analysis_method": "structural_analysis"
128
+ },
129
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
130
+ "timestamp": 1777250303.1900122,
131
+ "type": "onnx_check",
132
+ "rule_code": "S803"
133
+ },
134
+ {
135
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 272 has unusually dissimilar weights",
136
+ "severity": "info",
137
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
138
+ "details": {
139
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
140
+ "neuron_index": 272,
141
+ "max_similarity_to_others": 0.6721497178077698,
142
+ "weight_norm": 0.4996761083602905,
143
+ "total_outputs": 768,
144
+ "analysis_method": "structural_analysis"
145
+ },
146
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
147
+ "timestamp": 1777250303.1905012,
148
+ "type": "onnx_check",
149
+ "rule_code": "S803"
150
+ },
151
+ {
152
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 296 has unusually dissimilar weights",
153
+ "severity": "info",
154
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
155
+ "details": {
156
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
157
+ "neuron_index": 296,
158
+ "max_similarity_to_others": 0.6973440647125244,
159
+ "weight_norm": 0.5266340374946594,
160
+ "total_outputs": 768,
161
+ "analysis_method": "structural_analysis"
162
+ },
163
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
164
+ "timestamp": 1777250303.1910307,
165
+ "type": "onnx_check",
166
+ "rule_code": "S803"
167
+ },
168
+ {
169
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 329 has unusually dissimilar weights",
170
+ "severity": "info",
171
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
172
+ "details": {
173
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
174
+ "neuron_index": 329,
175
+ "max_similarity_to_others": 0.6732807159423828,
176
+ "weight_norm": 0.7019028663635254,
177
+ "total_outputs": 768,
178
+ "analysis_method": "structural_analysis"
179
+ },
180
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
181
+ "timestamp": 1777250303.191509,
182
+ "type": "onnx_check",
183
+ "rule_code": "S803"
184
+ },
185
+ {
186
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 344 has unusually dissimilar weights",
187
+ "severity": "info",
188
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
189
+ "details": {
190
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
191
+ "neuron_index": 344,
192
+ "max_similarity_to_others": 0.5913342237472534,
193
+ "weight_norm": 1.2862828969955444,
194
+ "total_outputs": 768,
195
+ "analysis_method": "structural_analysis"
196
+ },
197
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
198
+ "timestamp": 1777250303.1919985,
199
+ "type": "onnx_check",
200
+ "rule_code": "S803"
201
+ },
202
+ {
203
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 357 has unusually dissimilar weights",
204
+ "severity": "info",
205
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
206
+ "details": {
207
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
208
+ "neuron_index": 357,
209
+ "max_similarity_to_others": 0.6641792058944702,
210
+ "weight_norm": 0.5504754185676575,
211
+ "total_outputs": 768,
212
+ "analysis_method": "structural_analysis"
213
+ },
214
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
215
+ "timestamp": 1777250303.192473,
216
+ "type": "onnx_check",
217
+ "rule_code": "S803"
218
+ },
219
+ {
220
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 378 has unusually dissimilar weights",
221
+ "severity": "info",
222
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
223
+ "details": {
224
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
225
+ "neuron_index": 378,
226
+ "max_similarity_to_others": 0.5857805609703064,
227
+ "weight_norm": 0.5392732620239258,
228
+ "total_outputs": 768,
229
+ "analysis_method": "structural_analysis"
230
+ },
231
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
232
+ "timestamp": 1777250303.1929657,
233
+ "type": "onnx_check",
234
+ "rule_code": "S803"
235
+ },
236
+ {
237
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 426 has unusually dissimilar weights",
238
+ "severity": "info",
239
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
240
+ "details": {
241
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
242
+ "neuron_index": 426,
243
+ "max_similarity_to_others": 0.6619596481323242,
244
+ "weight_norm": 0.4695207178592682,
245
+ "total_outputs": 768,
246
+ "analysis_method": "structural_analysis"
247
+ },
248
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
249
+ "timestamp": 1777250303.1934395,
250
+ "type": "onnx_check",
251
+ "rule_code": "S803"
252
+ },
253
+ {
254
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 438 has unusually dissimilar weights",
255
+ "severity": "info",
256
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
257
+ "details": {
258
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
259
+ "neuron_index": 438,
260
+ "max_similarity_to_others": 0.6229392290115356,
261
+ "weight_norm": 0.43737903237342834,
262
+ "total_outputs": 768,
263
+ "analysis_method": "structural_analysis"
264
+ },
265
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
266
+ "timestamp": 1777250303.1939251,
267
+ "type": "onnx_check",
268
+ "rule_code": "S803"
269
+ },
270
+ {
271
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 455 has unusually dissimilar weights",
272
+ "severity": "info",
273
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
274
+ "details": {
275
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
276
+ "neuron_index": 455,
277
+ "max_similarity_to_others": 0.6861489415168762,
278
+ "weight_norm": 0.500296413898468,
279
+ "total_outputs": 768,
280
+ "analysis_method": "structural_analysis"
281
+ },
282
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
283
+ "timestamp": 1777250303.1944077,
284
+ "type": "onnx_check",
285
+ "rule_code": "S803"
286
+ },
287
+ {
288
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 481 has unusually dissimilar weights",
289
+ "severity": "info",
290
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
291
+ "details": {
292
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
293
+ "neuron_index": 481,
294
+ "max_similarity_to_others": 0.6357624530792236,
295
+ "weight_norm": 0.589555561542511,
296
+ "total_outputs": 768,
297
+ "analysis_method": "structural_analysis"
298
+ },
299
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
300
+ "timestamp": 1777250303.1949005,
301
+ "type": "onnx_check",
302
+ "rule_code": "S803"
303
+ },
304
+ {
305
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 487 has unusually dissimilar weights",
306
+ "severity": "info",
307
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
308
+ "details": {
309
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
310
+ "neuron_index": 487,
311
+ "max_similarity_to_others": 0.6871294975280762,
312
+ "weight_norm": 0.47944051027297974,
313
+ "total_outputs": 768,
314
+ "analysis_method": "structural_analysis"
315
+ },
316
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
317
+ "timestamp": 1777250303.195397,
318
+ "type": "onnx_check",
319
+ "rule_code": "S803"
320
+ },
321
+ {
322
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 544 has unusually dissimilar weights",
323
+ "severity": "info",
324
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
325
+ "details": {
326
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
327
+ "neuron_index": 544,
328
+ "max_similarity_to_others": 0.6773201823234558,
329
+ "weight_norm": 0.5820121169090271,
330
+ "total_outputs": 768,
331
+ "analysis_method": "structural_analysis"
332
+ },
333
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
334
+ "timestamp": 1777250303.195918,
335
+ "type": "onnx_check",
336
+ "rule_code": "S803"
337
+ },
338
+ {
339
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 719 has unusually dissimilar weights",
340
+ "severity": "info",
341
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
342
+ "details": {
343
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
344
+ "neuron_index": 719,
345
+ "max_similarity_to_others": 0.6987954378128052,
346
+ "weight_norm": 0.7778415083885193,
347
+ "total_outputs": 768,
348
+ "analysis_method": "structural_analysis"
349
+ },
350
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
351
+ "timestamp": 1777250303.1964004,
352
+ "type": "onnx_check",
353
+ "rule_code": "S803"
354
+ },
355
+ {
356
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 729 has unusually dissimilar weights",
357
+ "severity": "info",
358
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
359
+ "details": {
360
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
361
+ "neuron_index": 729,
362
+ "max_similarity_to_others": 0.6774657964706421,
363
+ "weight_norm": 0.6795850396156311,
364
+ "total_outputs": 768,
365
+ "analysis_method": "structural_analysis"
366
+ },
367
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
368
+ "timestamp": 1777250303.1969016,
369
+ "type": "onnx_check",
370
+ "rule_code": "S803"
371
+ },
372
+ {
373
+ "message": "Layer 'core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight' output neuron 739 has unusually dissimilar weights",
374
+ "severity": "info",
375
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
376
+ "details": {
377
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
378
+ "neuron_index": 739,
379
+ "max_similarity_to_others": 0.6594510674476624,
380
+ "weight_norm": 0.8022450804710388,
381
+ "total_outputs": 768,
382
+ "analysis_method": "structural_analysis"
383
+ },
384
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
385
+ "timestamp": 1777250303.1973808,
386
+ "type": "onnx_check",
387
+ "rule_code": "S803"
388
+ }
389
+ ],
390
+ "checks": [
391
+ {
392
+ "name": "Path Exists",
393
+ "status": "passed",
394
+ "message": "Path exists",
395
+ "location": "/opt/sas/model-gli-text/models/medium/README.md",
396
+ "details": {
397
+ "path": "/opt/sas/model-gli-text/models/medium/README.md"
398
+ },
399
+ "timestamp": 1777249824.745477
400
+ },
401
+ {
402
+ "name": "Path Readable",
403
+ "status": "passed",
404
+ "message": "Path is readable",
405
+ "location": "/opt/sas/model-gli-text/models/medium/README.md",
406
+ "details": {
407
+ "path": "/opt/sas/model-gli-text/models/medium/README.md"
408
+ },
409
+ "timestamp": 1777249824.7455053
410
+ },
411
+ {
412
+ "name": "File Type Validation",
413
+ "status": "passed",
414
+ "message": "File type validation passed",
415
+ "location": "/opt/sas/model-gli-text/models/medium/README.md",
416
+ "details": {},
417
+ "timestamp": 1777249824.7455347
418
+ },
419
+ {
420
+ "name": "Path Exists",
421
+ "status": "passed",
422
+ "message": "Path exists",
423
+ "location": "/opt/sas/model-gli-text/models/medium/gliner_config.json",
424
+ "details": {
425
+ "path": "/opt/sas/model-gli-text/models/medium/gliner_config.json"
426
+ },
427
+ "timestamp": 1777249824.7813027
428
+ },
429
+ {
430
+ "name": "Path Readable",
431
+ "status": "passed",
432
+ "message": "Path is readable",
433
+ "location": "/opt/sas/model-gli-text/models/medium/gliner_config.json",
434
+ "details": {
435
+ "path": "/opt/sas/model-gli-text/models/medium/gliner_config.json"
436
+ },
437
+ "timestamp": 1777249824.7813318
438
+ },
439
+ {
440
+ "name": "File Type Validation",
441
+ "status": "passed",
442
+ "message": "File type validation passed",
443
+ "location": "/opt/sas/model-gli-text/models/medium/gliner_config.json",
444
+ "details": {},
445
+ "timestamp": 1777249824.7813616
446
+ },
447
+ {
448
+ "name": "Model Name Policy Check",
449
+ "status": "passed",
450
+ "message": "Model Name Policy Check completed successfully",
451
+ "location": "/opt/sas/model-gli-text/models/medium/gliner_config.json",
452
+ "details": {
453
+ "component_count": 2
454
+ },
455
+ "timestamp": 1777249824.7826722
456
+ },
457
+ {
458
+ "name": "Path Exists",
459
+ "status": "passed",
460
+ "message": "Path exists",
461
+ "location": "/opt/sas/model-gli-text/models/medium/tokenizer_config.json",
462
+ "details": {
463
+ "path": "/opt/sas/model-gli-text/models/medium/tokenizer_config.json"
464
+ },
465
+ "timestamp": 1777249824.8342366
466
+ },
467
+ {
468
+ "name": "Path Readable",
469
+ "status": "passed",
470
+ "message": "Path is readable",
471
+ "location": "/opt/sas/model-gli-text/models/medium/tokenizer_config.json",
472
+ "details": {
473
+ "path": "/opt/sas/model-gli-text/models/medium/tokenizer_config.json"
474
+ },
475
+ "timestamp": 1777249824.83427
476
+ },
477
+ {
478
+ "name": "File Type Validation",
479
+ "status": "passed",
480
+ "message": "File type validation passed",
481
+ "location": "/opt/sas/model-gli-text/models/medium/tokenizer_config.json",
482
+ "details": {},
483
+ "timestamp": 1777249824.8343112
484
+ },
485
+ {
486
+ "name": "Template Extraction",
487
+ "status": "passed",
488
+ "message": "No Jinja2 templates found in file",
489
+ "location": "/opt/sas/model-gli-text/models/medium/tokenizer_config.json",
490
+ "details": {
491
+ "file_type": "tokenizer_config"
492
+ },
493
+ "timestamp": 1777249824.8348958
494
+ },
495
+ {
496
+ "name": "Path Exists",
497
+ "status": "passed",
498
+ "message": "Path exists",
499
+ "location": "/opt/sas/model-gli-text/models/medium/tokenizer.json",
500
+ "details": {
501
+ "path": "/opt/sas/model-gli-text/models/medium/tokenizer.json"
502
+ },
503
+ "timestamp": 1777249826.885802
504
+ },
505
+ {
506
+ "name": "Path Readable",
507
+ "status": "passed",
508
+ "message": "Path is readable",
509
+ "location": "/opt/sas/model-gli-text/models/medium/tokenizer.json",
510
+ "details": {
511
+ "path": "/opt/sas/model-gli-text/models/medium/tokenizer.json"
512
+ },
513
+ "timestamp": 1777249826.8858373
514
+ },
515
+ {
516
+ "name": "File Type Validation",
517
+ "status": "passed",
518
+ "message": "File type validation passed",
519
+ "location": "/opt/sas/model-gli-text/models/medium/tokenizer.json",
520
+ "details": {},
521
+ "timestamp": 1777249826.8858702
522
+ },
523
+ {
524
+ "name": "Template Extraction",
525
+ "status": "passed",
526
+ "message": "No Jinja2 templates found in file",
527
+ "location": "/opt/sas/model-gli-text/models/medium/tokenizer.json",
528
+ "details": {
529
+ "file_type": "tokenizer_config"
530
+ },
531
+ "timestamp": 1777249827.3324003
532
+ },
533
+ {
534
+ "name": "Path Exists",
535
+ "status": "passed",
536
+ "message": "Path exists",
537
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
538
+ "details": {
539
+ "path": "/opt/sas/model-gli-text/models/medium/model.onnx"
540
+ },
541
+ "timestamp": 1777249827.3575664
542
+ },
543
+ {
544
+ "name": "Path Readable",
545
+ "status": "passed",
546
+ "message": "Path is readable",
547
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
548
+ "details": {
549
+ "path": "/opt/sas/model-gli-text/models/medium/model.onnx"
550
+ },
551
+ "timestamp": 1777249827.3575964
552
+ },
553
+ {
554
+ "name": "File Type Validation",
555
+ "status": "passed",
556
+ "message": "File type validation passed",
557
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
558
+ "details": {},
559
+ "timestamp": 1777249827.3576233
560
+ },
561
+ {
562
+ "name": "File Integrity Hash",
563
+ "status": "passed",
564
+ "message": "File integrity hashes calculated",
565
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
566
+ "details": {
567
+ "md5": "67fab769e446bb64f1fc435a056bb656",
568
+ "sha256": "dfbf82b4c9b7cb8ebedb6e04bec068db7cfee3053cb90c7d4aa11b8a93edb46a",
569
+ "sha512": "4beeed401d13ff182de78ce3cc8829e45fba844b8952bdd7d4056c95fae5706216c57941b0793e7c51323af80d5d91ccf6f51ee8e2f9a2a2062ec1137cba6865",
570
+ "file_size": 835590772
571
+ },
572
+ "timestamp": 1777249831.6727397
573
+ },
574
+ {
575
+ "name": "JIT/Script Code Execution Detection",
576
+ "status": "passed",
577
+ "message": "No JIT/Script code execution risks detected",
578
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
579
+ "details": {},
580
+ "timestamp": 1777250298.1521664
581
+ },
582
+ {
583
+ "name": "Network Communication Detection",
584
+ "status": "passed",
585
+ "message": "No network communication patterns detected",
586
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
587
+ "details": {},
588
+ "timestamp": 1777250298.152354
589
+ },
590
+ {
591
+ "name": "Custom Operator Domain Check",
592
+ "status": "passed",
593
+ "message": "All operators use standard ONNX domains",
594
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
595
+ "details": {
596
+ "safe_nodes": 5171
597
+ },
598
+ "timestamp": 1777250298.1769123
599
+ },
600
+ {
601
+ "name": "Python Operator Detection",
602
+ "status": "passed",
603
+ "message": "No Python operators detected",
604
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
605
+ "details": {
606
+ "nodes_checked": 5171
607
+ },
608
+ "timestamp": 1777250298.1769633
609
+ },
610
+ {
611
+ "name": "Tensor Size Validation",
612
+ "status": "passed",
613
+ "message": "Tensor Size Validation completed successfully",
614
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx (tensor: core.token_rep_layer.bert_layer.model.embeddings.word_embeddings.weight)",
615
+ "details": {
616
+ "component_count": 217
617
+ },
618
+ "timestamp": 1777250299.9097366
619
+ },
620
+ {
621
+ "name": "Weight Distribution Analysis Coverage",
622
+ "status": "failed",
623
+ "message": "Weight distribution analysis skipped one or more eligible ONNX initializers",
624
+ "severity": "info",
625
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
626
+ "details": {
627
+ "scan_outcome_reason": "onnx_weight_distribution_analysis_incomplete",
628
+ "coverage_gap": "partial_initializer_coverage",
629
+ "eligible_initializers": 85,
630
+ "analyzed_initializers": 84,
631
+ "external_initializers_skipped": 0,
632
+ "oversized_initializers_skipped": 1,
633
+ "extraction_failures": 0,
634
+ "max_array_size": 104857600
635
+ },
636
+ "rule_code": "S902",
637
+ "timestamp": 1777250300.2641213
638
+ },
639
+ {
640
+ "name": "Weight Distribution Anomaly Detection",
641
+ "status": "failed",
642
+ "message": "Weight Distribution Anomaly Detection found 21 issues",
643
+ "severity": "info",
644
+ "location": "/opt/sas/model-gli-text/models/medium/model.onnx",
645
+ "details": {
646
+ "component_count": 21,
647
+ "findings": [
648
+ {
649
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
650
+ "neuron_index": 7,
651
+ "max_similarity_to_others": 0.6353273391723633,
652
+ "weight_norm": 5.372602462768555,
653
+ "total_outputs": 768,
654
+ "analysis_method": "structural_analysis"
655
+ },
656
+ {
657
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
658
+ "neuron_index": 48,
659
+ "max_similarity_to_others": 0.6696844100952148,
660
+ "weight_norm": 0.46152299642562866,
661
+ "total_outputs": 768,
662
+ "analysis_method": "structural_analysis"
663
+ },
664
+ {
665
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
666
+ "neuron_index": 117,
667
+ "max_similarity_to_others": 0.634371280670166,
668
+ "weight_norm": 0.39997434616088867,
669
+ "total_outputs": 768,
670
+ "analysis_method": "structural_analysis"
671
+ },
672
+ {
673
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
674
+ "neuron_index": 213,
675
+ "max_similarity_to_others": 0.6954333186149597,
676
+ "weight_norm": 0.5225619673728943,
677
+ "total_outputs": 768,
678
+ "analysis_method": "structural_analysis"
679
+ },
680
+ {
681
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
682
+ "neuron_index": 248,
683
+ "max_similarity_to_others": 0.6133643388748169,
684
+ "weight_norm": 0.47258928418159485,
685
+ "total_outputs": 768,
686
+ "analysis_method": "structural_analysis"
687
+ },
688
+ {
689
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
690
+ "neuron_index": 266,
691
+ "max_similarity_to_others": 0.649356484413147,
692
+ "weight_norm": 0.683588981628418,
693
+ "total_outputs": 768,
694
+ "analysis_method": "structural_analysis"
695
+ },
696
+ {
697
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
698
+ "neuron_index": 272,
699
+ "max_similarity_to_others": 0.6721497178077698,
700
+ "weight_norm": 0.4996761083602905,
701
+ "total_outputs": 768,
702
+ "analysis_method": "structural_analysis"
703
+ },
704
+ {
705
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
706
+ "neuron_index": 296,
707
+ "max_similarity_to_others": 0.6973440647125244,
708
+ "weight_norm": 0.5266340374946594,
709
+ "total_outputs": 768,
710
+ "analysis_method": "structural_analysis"
711
+ },
712
+ {
713
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
714
+ "neuron_index": 329,
715
+ "max_similarity_to_others": 0.6732807159423828,
716
+ "weight_norm": 0.7019028663635254,
717
+ "total_outputs": 768,
718
+ "analysis_method": "structural_analysis"
719
+ },
720
+ {
721
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
722
+ "neuron_index": 344,
723
+ "max_similarity_to_others": 0.5913342237472534,
724
+ "weight_norm": 1.2862828969955444,
725
+ "total_outputs": 768,
726
+ "analysis_method": "structural_analysis"
727
+ },
728
+ {
729
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
730
+ "neuron_index": 357,
731
+ "max_similarity_to_others": 0.6641792058944702,
732
+ "weight_norm": 0.5504754185676575,
733
+ "total_outputs": 768,
734
+ "analysis_method": "structural_analysis"
735
+ },
736
+ {
737
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
738
+ "neuron_index": 378,
739
+ "max_similarity_to_others": 0.5857805609703064,
740
+ "weight_norm": 0.5392732620239258,
741
+ "total_outputs": 768,
742
+ "analysis_method": "structural_analysis"
743
+ },
744
+ {
745
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
746
+ "neuron_index": 426,
747
+ "max_similarity_to_others": 0.6619596481323242,
748
+ "weight_norm": 0.4695207178592682,
749
+ "total_outputs": 768,
750
+ "analysis_method": "structural_analysis"
751
+ },
752
+ {
753
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
754
+ "neuron_index": 438,
755
+ "max_similarity_to_others": 0.6229392290115356,
756
+ "weight_norm": 0.43737903237342834,
757
+ "total_outputs": 768,
758
+ "analysis_method": "structural_analysis"
759
+ },
760
+ {
761
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
762
+ "neuron_index": 455,
763
+ "max_similarity_to_others": 0.6861489415168762,
764
+ "weight_norm": 0.500296413898468,
765
+ "total_outputs": 768,
766
+ "analysis_method": "structural_analysis"
767
+ },
768
+ {
769
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
770
+ "neuron_index": 481,
771
+ "max_similarity_to_others": 0.6357624530792236,
772
+ "weight_norm": 0.589555561542511,
773
+ "total_outputs": 768,
774
+ "analysis_method": "structural_analysis"
775
+ },
776
+ {
777
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
778
+ "neuron_index": 487,
779
+ "max_similarity_to_others": 0.6871294975280762,
780
+ "weight_norm": 0.47944051027297974,
781
+ "total_outputs": 768,
782
+ "analysis_method": "structural_analysis"
783
+ },
784
+ {
785
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
786
+ "neuron_index": 544,
787
+ "max_similarity_to_others": 0.6773201823234558,
788
+ "weight_norm": 0.5820121169090271,
789
+ "total_outputs": 768,
790
+ "analysis_method": "structural_analysis"
791
+ },
792
+ {
793
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
794
+ "neuron_index": 719,
795
+ "max_similarity_to_others": 0.6987954378128052,
796
+ "weight_norm": 0.7778415083885193,
797
+ "total_outputs": 768,
798
+ "analysis_method": "structural_analysis"
799
+ },
800
+ {
801
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
802
+ "neuron_index": 729,
803
+ "max_similarity_to_others": 0.6774657964706421,
804
+ "weight_norm": 0.6795850396156311,
805
+ "total_outputs": 768,
806
+ "analysis_method": "structural_analysis"
807
+ },
808
+ {
809
+ "layer": "core.token_rep_layer.bert_layer.model.encoder.rel_embeddings.weight",
810
+ "neuron_index": 739,
811
+ "max_similarity_to_others": 0.6594510674476624,
812
+ "weight_norm": 0.8022450804710388,
813
+ "total_outputs": 768,
814
+ "analysis_method": "structural_analysis"
815
+ }
816
+ ]
817
+ },
818
+ "why": "Neurons with weight patterns completely unlike others in the same layer are uncommon in standard training. This dissimilarity (measured by cosine similarity below threshold) may indicate injected functionality or training irregularities.",
819
+ "timestamp": 1777250303.197376
820
+ }
821
+ ],
822
+ "files_scanned": 6,
823
+ "assets": [
824
+ {
825
+ "path": "/opt/sas/model-gli-text/models/medium/README.md",
826
+ "type": "metadata",
827
+ "size": 1591
828
+ },
829
+ {
830
+ "path": "/opt/sas/model-gli-text/models/medium/gliner_config.json",
831
+ "type": "manifest",
832
+ "size": 2178,
833
+ "keys": [
834
+ "class_token_index",
835
+ "dropout",
836
+ "embed_ent_token",
837
+ "encoder_config",
838
+ "ent_token",
839
+ "fine_tune",
840
+ "fuse_layers",
841
+ "has_rnn",
842
+ "hidden_size",
843
+ "label_smoothing",
844
+ "loss_alpha",
845
+ "loss_gamma",
846
+ "loss_reduction",
847
+ "max_len",
848
+ "max_neg_type_ratio",
849
+ "max_types",
850
+ "max_width",
851
+ "model_name",
852
+ "model_type",
853
+ "name",
854
+ "neg_spans_ratio",
855
+ "num_post_fusion_layers",
856
+ "num_rnn_layers",
857
+ "post_fusion_schema",
858
+ "random_drop",
859
+ "represent_spans",
860
+ "sep_token",
861
+ "shuffle_types",
862
+ "span_loss_coef",
863
+ "span_mode",
864
+ "subtoken_pooling",
865
+ "token_loss_coef",
866
+ "transformers_version",
867
+ "vocab_size",
868
+ "words_splitter_type"
869
+ ]
870
+ },
871
+ {
872
+ "path": "/opt/sas/model-gli-text/models/medium/tokenizer_config.json",
873
+ "type": "jinja2_template",
874
+ "size": 502
875
+ },
876
+ {
877
+ "path": "/opt/sas/model-gli-text/models/medium/tokenizer.json",
878
+ "type": "jinja2_template",
879
+ "size": 8332739
880
+ },
881
+ {
882
+ "path": "/opt/sas/model-gli-text/models/medium/model.onnx",
883
+ "type": "onnx",
884
+ "size": 835590772
885
+ },
886
+ {
887
+ "path": "/opt/sas/model-gli-text/models/medium/LICENSE",
888
+ "type": "unknown"
889
+ }
890
+ ],
891
+ "has_errors": false,
892
+ "scanner_names": [
893
+ "metadata",
894
+ "manifest",
895
+ "jinja2_template",
896
+ "onnx"
897
+ ],
898
+ "file_metadata": {
899
+ "/opt/sas/model-gli-text/models/medium/README.md": {
900
+ "file_size": 1591,
901
+ "license_info": [
902
+ {
903
+ "spdx_id": "Apache-2.0",
904
+ "name": "Apache License 2.0",
905
+ "confidence": 0.8,
906
+ "source": "file_header",
907
+ "commercial_allowed": true
908
+ }
909
+ ],
910
+ "copyright_notices": [],
911
+ "license_files_nearby": [
912
+ "/opt/sas/model-gli-text/models/medium/LICENSE"
913
+ ],
914
+ "is_dataset": false,
915
+ "is_model": false,
916
+ "risk_score": 0.0,
917
+ "scan_timestamp": 1777249824.7508566,
918
+ "content_hash": "7999f932aa2777fb2d12a1e13ef146c5ce362bf185c600313c161d8f1618d4be"
919
+ },
920
+ "/opt/sas/model-gli-text/models/medium/gliner_config.json": {
921
+ "file_size": 2178,
922
+ "license_info": [],
923
+ "copyright_notices": [],
924
+ "license_files_nearby": [
925
+ "/opt/sas/model-gli-text/models/medium/LICENSE"
926
+ ],
927
+ "is_dataset": true,
928
+ "is_model": false,
929
+ "risk_score": 0.0,
930
+ "scan_timestamp": 1777249824.7844925,
931
+ "root_type": "dict",
932
+ "keys": [
933
+ "class_token_index",
934
+ "dropout",
935
+ "embed_ent_token",
936
+ "encoder_config",
937
+ "ent_token",
938
+ "fine_tune",
939
+ "fuse_layers",
940
+ "has_rnn",
941
+ "hidden_size",
942
+ "label_smoothing",
943
+ "loss_alpha",
944
+ "loss_gamma",
945
+ "loss_reduction",
946
+ "max_len",
947
+ "max_neg_type_ratio",
948
+ "max_types",
949
+ "max_width",
950
+ "model_name",
951
+ "model_type",
952
+ "name",
953
+ "neg_spans_ratio",
954
+ "num_post_fusion_layers",
955
+ "num_rnn_layers",
956
+ "post_fusion_schema",
957
+ "random_drop",
958
+ "represent_spans",
959
+ "sep_token",
960
+ "shuffle_types",
961
+ "span_loss_coef",
962
+ "span_mode",
963
+ "subtoken_pooling",
964
+ "token_loss_coef",
965
+ "transformers_version",
966
+ "vocab_size",
967
+ "words_splitter_type"
968
+ ],
969
+ "content_hash": "f538f4dc5ac579efed802ea637a413b8038d621f62f6acfb6f74e89eb7514be4"
970
+ },
971
+ "/opt/sas/model-gli-text/models/medium/tokenizer_config.json": {
972
+ "file_size": 502,
973
+ "ml_context": {
974
+ "frameworks": {},
975
+ "overall_confidence": 0.0,
976
+ "is_ml_content": false,
977
+ "detected_patterns": [],
978
+ "optimization_hints": [],
979
+ "file_type": "tokenizer_config",
980
+ "is_tokenizer": true,
981
+ "confidence": 2
982
+ },
983
+ "license_info": [],
984
+ "copyright_notices": [],
985
+ "license_files_nearby": [
986
+ "/opt/sas/model-gli-text/models/medium/LICENSE"
987
+ ],
988
+ "is_dataset": true,
989
+ "is_model": false,
990
+ "risk_score": 0.0,
991
+ "scan_timestamp": 1777249824.836328,
992
+ "content_hash": "818c97fbc7f072800ccc379fd72e1e1b2506d30d7fd5bee8a5e4c8bd61bd805c"
993
+ },
994
+ "/opt/sas/model-gli-text/models/medium/tokenizer.json": {
995
+ "file_size": 8332739,
996
+ "ml_context": {
997
+ "frameworks": {},
998
+ "overall_confidence": 0.0,
999
+ "is_ml_content": false,
1000
+ "detected_patterns": [],
1001
+ "optimization_hints": [],
1002
+ "file_type": "tokenizer_config",
1003
+ "is_tokenizer": true,
1004
+ "confidence": 2
1005
+ },
1006
+ "license_info": [],
1007
+ "copyright_notices": [],
1008
+ "license_files_nearby": [
1009
+ "/opt/sas/model-gli-text/models/medium/LICENSE"
1010
+ ],
1011
+ "is_dataset": true,
1012
+ "is_model": false,
1013
+ "risk_score": 0.0,
1014
+ "scan_timestamp": 1777249827.3364515,
1015
+ "content_hash": "08bb5853718f4a829fa9ce773d7984f7f3f6a7073fdc82a07a382675c5061ba6"
1016
+ },
1017
+ "/opt/sas/model-gli-text/models/medium/model.onnx": {
1018
+ "file_size": 835590772,
1019
+ "file_hashes": {
1020
+ "md5": "67fab769e446bb64f1fc435a056bb656",
1021
+ "sha256": "dfbf82b4c9b7cb8ebedb6e04bec068db7cfee3053cb90c7d4aa11b8a93edb46a",
1022
+ "sha512": "4beeed401d13ff182de78ce3cc8829e45fba844b8952bdd7d4056c95fae5706216c57941b0793e7c51323af80d5d91ccf6f51ee8e2f9a2a2062ec1137cba6865"
1023
+ },
1024
+ "license_info": [],
1025
+ "copyright_notices": [],
1026
+ "license_files_nearby": [
1027
+ "/opt/sas/model-gli-text/models/medium/LICENSE"
1028
+ ],
1029
+ "is_dataset": false,
1030
+ "is_model": true,
1031
+ "risk_score": 0.0,
1032
+ "scan_timestamp": 1777250303.7373185,
1033
+ "ir_version": 7,
1034
+ "producer_name": "pytorch",
1035
+ "node_count": 5171,
1036
+ "scan_outcome": "inconclusive",
1037
+ "scan_outcome_reasons": [
1038
+ "onnx_weight_distribution_analysis_incomplete"
1039
+ ],
1040
+ "layers_analyzed": 84,
1041
+ "anomalies_found": 21,
1042
+ "content_hash": "dfbf82b4c9b7cb8ebedb6e04bec068db7cfee3053cb90c7d4aa11b8a93edb46a"
1043
+ },
1044
+ "/opt/sas/model-gli-text/models/medium/LICENSE": {
1045
+ "license_info": [
1046
+ {
1047
+ "spdx_id": "Apache-2.0",
1048
+ "name": "Apache License 2.0",
1049
+ "confidence": 0.8,
1050
+ "source": "file_header",
1051
+ "commercial_allowed": true
1052
+ }
1053
+ ],
1054
+ "copyright_notices": [],
1055
+ "license_files_nearby": [
1056
+ "/opt/sas/model-gli-text/models/medium/LICENSE"
1057
+ ],
1058
+ "is_dataset": false,
1059
+ "is_model": false,
1060
+ "risk_score": 0.0,
1061
+ "scan_timestamp": 1777250303.7449977,
1062
+ "content_hash": "cfc7749b96f63bd31c3c42b5c471bf756814053e847c10f3eb003417bc523d30"
1063
+ }
1064
+ },
1065
+ "content_hash": "8f06b3046d0d86ac35db3b5469e20a6e85aeb41b5b2e384e97535b79a654d23c",
1066
+ "start_time": 1777249823.545908,
1067
+ "duration": 480.208021402359,
1068
+ "total_checks": 24,
1069
+ "passed_checks": 24,
1070
+ "failed_checks": 0,
1071
+ "success": false
1072
+ }
1073
+ }
tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json CHANGED
@@ -1,67 +1,12 @@
1
  {
2
- "added_tokens_decoder": {
3
- "0": {
4
- "content": "[PAD]",
5
- "lstrip": false,
6
- "normalized": false,
7
- "rstrip": false,
8
- "single_word": false,
9
- "special": true
10
- },
11
- "1": {
12
- "content": "[CLS]",
13
- "lstrip": false,
14
- "normalized": false,
15
- "rstrip": false,
16
- "single_word": false,
17
- "special": true
18
- },
19
- "2": {
20
- "content": "[SEP]",
21
- "lstrip": false,
22
- "normalized": false,
23
- "rstrip": false,
24
- "single_word": false,
25
- "special": true
26
- },
27
- "3": {
28
- "content": "[UNK]",
29
- "lstrip": false,
30
- "normalized": true,
31
- "rstrip": false,
32
- "single_word": false,
33
- "special": true
34
- },
35
- "128000": {
36
- "content": "[MASK]",
37
- "lstrip": false,
38
- "normalized": false,
39
- "rstrip": false,
40
- "single_word": false,
41
- "special": true
42
- },
43
- "128001": {
44
- "content": "<<ENT>>",
45
- "lstrip": false,
46
- "normalized": true,
47
- "rstrip": false,
48
- "single_word": false,
49
- "special": false
50
- },
51
- "128002": {
52
- "content": "<<SEP>>",
53
- "lstrip": false,
54
- "normalized": true,
55
- "rstrip": false,
56
- "single_word": false,
57
- "special": false
58
- }
59
- },
60
  "bos_token": "[CLS]",
61
  "clean_up_tokenization_spaces": true,
62
  "cls_token": "[CLS]",
63
  "do_lower_case": false,
64
  "eos_token": "[SEP]",
 
65
  "mask_token": "[MASK]",
66
  "model_max_length": 1000000000000000019884624838656,
67
  "pad_token": "[PAD]",
@@ -69,6 +14,7 @@
69
  "sp_model_kwargs": {},
70
  "split_by_punct": false,
71
  "tokenizer_class": "DebertaV2Tokenizer",
 
72
  "unk_token": "[UNK]",
73
  "vocab_type": "spm"
74
  }
 
1
  {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4
  "bos_token": "[CLS]",
5
  "clean_up_tokenization_spaces": true,
6
  "cls_token": "[CLS]",
7
  "do_lower_case": false,
8
  "eos_token": "[SEP]",
9
+ "is_local": true,
10
  "mask_token": "[MASK]",
11
  "model_max_length": 1000000000000000019884624838656,
12
  "pad_token": "[PAD]",
 
14
  "sp_model_kwargs": {},
15
  "split_by_punct": false,
16
  "tokenizer_class": "DebertaV2Tokenizer",
17
+ "unk_id": 3,
18
  "unk_token": "[UNK]",
19
  "vocab_type": "spm"
20
  }