Token Classification
Transformers
Safetensors
lfm2
liquid
lfm2.5
bidirectional
masked-lm
encoder
pii
ner
privacy
multilingual
custom_code
Instructions to use LiquidAI/LFM2.5-Encoder-350M-PII-Detector with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use LiquidAI/LFM2.5-Encoder-350M-PII-Detector with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="LiquidAI/LFM2.5-Encoder-350M-PII-Detector", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("LiquidAI/LFM2.5-Encoder-350M-PII-Detector", trust_remote_code=True) model = AutoModelForTokenClassification.from_pretrained("LiquidAI/LFM2.5-Encoder-350M-PII-Detector", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
v4: +v7 proxy in-distribution (gate F1 .60->.72), 32 types, #1 on all 4 clean held
Browse files- README.md +20 -6
- config.json +200 -176
- leaderboard.png +2 -2
- model.safetensors +2 -2
README.md
CHANGED
|
@@ -54,7 +54,7 @@ _clean · synthetic · tests PII-vs-NER discrimination_
|
|
| 54 |
|
| 55 |
| Model | P | R | F1 | F2 |
|
| 56 |
|---|---|---|---|---|
|
| 57 |
-
| trained:pii-detect-
|
| 58 |
| regex baseline | 0.259 | 0.579 | 0.358 | 0.464 |
|
| 59 |
| VAGOsolutions/SauerkrautLM-GLiNER | 0.171 | 0.785 | 0.280 | 0.456 |
|
| 60 |
| openai/privacy-filter | 0.169 | 0.595 | 0.264 | 0.396 |
|
|
@@ -67,7 +67,7 @@ _clean · synthetic English_
|
|
| 67 |
|
| 68 |
| Model | P | R | F1 | F2 |
|
| 69 |
|---|---|---|---|---|
|
| 70 |
-
| trained:pii-detect-
|
| 71 |
| LiquidAI demo (LFM2.5-PII-Demo) | 0.887 | 0.735 | 0.804 | 0.761 |
|
| 72 |
| OpenMed/privacy-filter-nemotron | 0.762 | 0.773 | 0.767 | 0.770 |
|
| 73 |
| VAGOsolutions/SauerkrautLM-GLiNER | 0.605 | 0.734 | 0.663 | 0.704 |
|
|
@@ -80,7 +80,7 @@ _clean · human gold · English legal (ECHR)_
|
|
| 80 |
|
| 81 |
| Model | P | R | F1 | F2 |
|
| 82 |
|---|---|---|---|---|
|
| 83 |
-
| trained:pii-detect-
|
| 84 |
| VAGOsolutions/SauerkrautLM-GLiNER | 0.843 | 0.576 | 0.685 | 0.615 |
|
| 85 |
| OpenMed/privacy-filter-nemotron | 0.728 | 0.625 | 0.673 | 0.643 |
|
| 86 |
| openai/privacy-filter | 0.933 | 0.383 | 0.543 | 0.434 |
|
|
@@ -93,7 +93,7 @@ _clean · human gold · multilingual EU legal_
|
|
| 93 |
|
| 94 |
| Model | P | R | F1 | F2 |
|
| 95 |
|---|---|---|---|---|
|
| 96 |
-
| trained:pii-detect-
|
| 97 |
| VAGOsolutions/SauerkrautLM-GLiNER | 0.487 | 0.258 | 0.337 | 0.285 |
|
| 98 |
| iiiorg/piiranha-v1 | 0.465 | 0.181 | 0.261 | 0.206 |
|
| 99 |
| openai/privacy-filter | 0.372 | 0.186 | 0.248 | 0.207 |
|
|
@@ -112,7 +112,7 @@ _⚠ in-distribution for OpenMed-nemotron & nvidia GLiNER (read as ceilings)_
|
|
| 112 |
| LiquidAI demo (LFM2.5-PII-Demo) | 0.894 | 0.518 | 0.656 | 0.566 |
|
| 113 |
| iiiorg/piiranha-v1 | 0.841 | 0.538 | 0.656 | 0.580 |
|
| 114 |
| VAGOsolutions/SauerkrautLM-GLiNER | 0.707 | 0.583 | 0.639 | 0.604 |
|
| 115 |
-
| trained:pii-detect-
|
| 116 |
| openai/privacy-filter | 0.699 | 0.484 | 0.572 | 0.516 |
|
| 117 |
| regex baseline | 0.645 | 0.227 | 0.335 | 0.260 |
|
| 118 |
|
|
@@ -125,8 +125,8 @@ _⚠ in OUR backbone's pretraining — contaminated for us; reference only_
|
|
| 125 |
| LiquidAI demo (LFM2.5-PII-Demo) | 0.533 | 0.464 | 0.496 | 0.476 |
|
| 126 |
| VAGOsolutions/SauerkrautLM-GLiNER | 0.403 | 0.610 | 0.485 | 0.553 |
|
| 127 |
| OpenMed/privacy-filter-nemotron | 0.341 | 0.541 | 0.418 | 0.484 |
|
|
|
|
| 128 |
| openai/privacy-filter | 0.316 | 0.492 | 0.385 | 0.443 |
|
| 129 |
-
| trained:pii-detect-v3a | 0.339 | 0.442 | 0.384 | 0.416 |
|
| 130 |
| regex baseline | 0.269 | 0.147 | 0.190 | 0.162 |
|
| 131 |
|
| 132 |
|
|
@@ -134,3 +134,17 @@ _⚠ in OUR backbone's pretraining — contaminated for us; reference only_
|
|
| 134 |
- **Strength:** best-in-class on **SPY** (PII-vs-NER discrimination) with the highest recall of any model tested.
|
| 135 |
- **Gap:** legal-domain benchmarks (TAB/MAPA) — high precision but low recall; targeted for the next data iteration.
|
| 136 |
- Benchmarks use differing entity taxonomies; our fine-grained types are mapped to each via an identity/crosswalk, so type-level F1 understates raw detection (binary detection is higher).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 54 |
|
| 55 |
| Model | P | R | F1 | F2 |
|
| 56 |
|---|---|---|---|---|
|
| 57 |
+
| trained:pii-detect-v4 | 0.284 | 0.706 | 0.405 | 0.544 |
|
| 58 |
| regex baseline | 0.259 | 0.579 | 0.358 | 0.464 |
|
| 59 |
| VAGOsolutions/SauerkrautLM-GLiNER | 0.171 | 0.785 | 0.280 | 0.456 |
|
| 60 |
| openai/privacy-filter | 0.169 | 0.595 | 0.264 | 0.396 |
|
|
|
|
| 67 |
|
| 68 |
| Model | P | R | F1 | F2 |
|
| 69 |
|---|---|---|---|---|
|
| 70 |
+
| trained:pii-detect-v4 | 0.770 | 0.852 | 0.809 | 0.834 |
|
| 71 |
| LiquidAI demo (LFM2.5-PII-Demo) | 0.887 | 0.735 | 0.804 | 0.761 |
|
| 72 |
| OpenMed/privacy-filter-nemotron | 0.762 | 0.773 | 0.767 | 0.770 |
|
| 73 |
| VAGOsolutions/SauerkrautLM-GLiNER | 0.605 | 0.734 | 0.663 | 0.704 |
|
|
|
|
| 80 |
|
| 81 |
| Model | P | R | F1 | F2 |
|
| 82 |
|---|---|---|---|---|
|
| 83 |
+
| trained:pii-detect-v4 | 0.860 | 0.782 | 0.819 | 0.797 |
|
| 84 |
| VAGOsolutions/SauerkrautLM-GLiNER | 0.843 | 0.576 | 0.685 | 0.615 |
|
| 85 |
| OpenMed/privacy-filter-nemotron | 0.728 | 0.625 | 0.673 | 0.643 |
|
| 86 |
| openai/privacy-filter | 0.933 | 0.383 | 0.543 | 0.434 |
|
|
|
|
| 93 |
|
| 94 |
| Model | P | R | F1 | F2 |
|
| 95 |
|---|---|---|---|---|
|
| 96 |
+
| trained:pii-detect-v4 | 0.445 | 0.541 | 0.488 | 0.518 |
|
| 97 |
| VAGOsolutions/SauerkrautLM-GLiNER | 0.487 | 0.258 | 0.337 | 0.285 |
|
| 98 |
| iiiorg/piiranha-v1 | 0.465 | 0.181 | 0.261 | 0.206 |
|
| 99 |
| openai/privacy-filter | 0.372 | 0.186 | 0.248 | 0.207 |
|
|
|
|
| 112 |
| LiquidAI demo (LFM2.5-PII-Demo) | 0.894 | 0.518 | 0.656 | 0.566 |
|
| 113 |
| iiiorg/piiranha-v1 | 0.841 | 0.538 | 0.656 | 0.580 |
|
| 114 |
| VAGOsolutions/SauerkrautLM-GLiNER | 0.707 | 0.583 | 0.639 | 0.604 |
|
| 115 |
+
| trained:pii-detect-v4 | 0.712 | 0.538 | 0.613 | 0.566 |
|
| 116 |
| openai/privacy-filter | 0.699 | 0.484 | 0.572 | 0.516 |
|
| 117 |
| regex baseline | 0.645 | 0.227 | 0.335 | 0.260 |
|
| 118 |
|
|
|
|
| 125 |
| LiquidAI demo (LFM2.5-PII-Demo) | 0.533 | 0.464 | 0.496 | 0.476 |
|
| 126 |
| VAGOsolutions/SauerkrautLM-GLiNER | 0.403 | 0.610 | 0.485 | 0.553 |
|
| 127 |
| OpenMed/privacy-filter-nemotron | 0.341 | 0.541 | 0.418 | 0.484 |
|
| 128 |
+
| trained:pii-detect-v4 | 0.330 | 0.488 | 0.394 | 0.445 |
|
| 129 |
| openai/privacy-filter | 0.316 | 0.492 | 0.385 | 0.443 |
|
|
|
|
| 130 |
| regex baseline | 0.269 | 0.147 | 0.190 | 0.162 |
|
| 131 |
|
| 132 |
|
|
|
|
| 134 |
- **Strength:** best-in-class on **SPY** (PII-vs-NER discrimination) with the highest recall of any model tested.
|
| 135 |
- **Gap:** legal-domain benchmarks (TAB/MAPA) — high precision but low recall; targeted for the next data iteration.
|
| 136 |
- Benchmarks use differing entity taxonomies; our fine-grained types are mapped to each via an identity/crosswalk, so type-level F1 understates raw detection (binary detection is higher).
|
| 137 |
+
|
| 138 |
+
## ShieldFlow proxy-traffic product gate (native ShieldFlow taxonomy, partial)
|
| 139 |
+
|
| 140 |
+
Held-out 10% of the v7 proxy corpus (enterprise pastes); trained on the same 90% the
|
| 141 |
+
incumbent did, evaluated on the identical untouched holdout.
|
| 142 |
+
|
| 143 |
+
| dataset | ours (v4) P / R / F1 | q4 incumbent |
|
| 144 |
+
|---|---|---|
|
| 145 |
+
| proxy_traffic_holdout | 0.749 / 0.683 / **0.714** | F1 0.878 / R 0.982 |
|
| 146 |
+
| proxy_traffic_holdout_xlong | 0.688 / 0.753 / **0.719** | F1 0.884 / R 0.940 |
|
| 147 |
+
|
| 148 |
+
v4 adds the v7 train split + 3 native ShieldFlow types (`developer.login_credentials`,
|
| 149 |
+
`developer.device_id`, `healthcare.health_plan_id`). Recall gap to the in-distribution
|
| 150 |
+
incumbent remains; tuning the v7 mix weight is the lever.
|
config.json
CHANGED
|
@@ -56,94 +56,106 @@
|
|
| 56 |
"26": "I-credential.password",
|
| 57 |
"27": "E-credential.password",
|
| 58 |
"28": "S-credential.password",
|
| 59 |
-
"29": "B-
|
| 60 |
-
"30": "I-
|
| 61 |
-
"31": "E-
|
| 62 |
-
"32": "S-
|
| 63 |
-
"33": "B-
|
| 64 |
-
"34": "I-
|
| 65 |
-
"35": "E-
|
| 66 |
-
"36": "S-
|
| 67 |
-
"37": "B-
|
| 68 |
-
"38": "I-
|
| 69 |
-
"39": "E-
|
| 70 |
-
"40": "S-
|
| 71 |
-
"41": "B-financial.
|
| 72 |
-
"42": "I-financial.
|
| 73 |
-
"43": "E-financial.
|
| 74 |
-
"44": "S-financial.
|
| 75 |
-
"45": "B-financial.
|
| 76 |
-
"46": "I-financial.
|
| 77 |
-
"47": "E-financial.
|
| 78 |
-
"48": "S-financial.
|
| 79 |
-
"49": "B-
|
| 80 |
-
"50": "I-
|
| 81 |
-
"51": "E-
|
| 82 |
-
"52": "S-
|
| 83 |
-
"53": "B-
|
| 84 |
-
"54": "I-
|
| 85 |
-
"55": "E-
|
| 86 |
-
"56": "S-
|
| 87 |
-
"57": "B-healthcare.
|
| 88 |
-
"58": "I-healthcare.
|
| 89 |
-
"59": "E-healthcare.
|
| 90 |
-
"60": "S-healthcare.
|
| 91 |
-
"61": "B-
|
| 92 |
-
"62": "I-
|
| 93 |
-
"63": "E-
|
| 94 |
-
"64": "S-
|
| 95 |
-
"65": "B-
|
| 96 |
-
"66": "I-
|
| 97 |
-
"67": "E-
|
| 98 |
-
"68": "S-
|
| 99 |
-
"69": "B-
|
| 100 |
-
"70": "I-
|
| 101 |
-
"71": "E-
|
| 102 |
-
"72": "S-
|
| 103 |
-
"73": "B-identity.
|
| 104 |
-
"74": "I-identity.
|
| 105 |
-
"75": "E-identity.
|
| 106 |
-
"76": "S-identity.
|
| 107 |
-
"77": "B-identity.
|
| 108 |
-
"78": "I-identity.
|
| 109 |
-
"79": "E-identity.
|
| 110 |
-
"80": "S-identity.
|
| 111 |
-
"81": "B-identity.
|
| 112 |
-
"82": "I-identity.
|
| 113 |
-
"83": "E-identity.
|
| 114 |
-
"84": "S-identity.
|
| 115 |
-
"85": "B-identity.
|
| 116 |
-
"86": "I-identity.
|
| 117 |
-
"87": "E-identity.
|
| 118 |
-
"88": "S-identity.
|
| 119 |
-
"89": "B-
|
| 120 |
-
"90": "I-
|
| 121 |
-
"91": "E-
|
| 122 |
-
"92": "S-
|
| 123 |
-
"93": "B-
|
| 124 |
-
"94": "I-
|
| 125 |
-
"95": "E-
|
| 126 |
-
"96": "S-
|
| 127 |
-
"97": "B-
|
| 128 |
-
"98": "I-
|
| 129 |
-
"99": "E-
|
| 130 |
-
"100": "S-
|
| 131 |
-
"101": "B-
|
| 132 |
-
"102": "I-
|
| 133 |
-
"103": "E-
|
| 134 |
-
"104": "S-
|
| 135 |
-
"105": "B-
|
| 136 |
-
"106": "I-
|
| 137 |
-
"107": "E-
|
| 138 |
-
"108": "S-
|
| 139 |
-
"109": "B-
|
| 140 |
-
"110": "I-
|
| 141 |
-
"111": "E-
|
| 142 |
-
"112": "S-
|
| 143 |
-
"113": "B-
|
| 144 |
-
"114": "I-
|
| 145 |
-
"115": "E-
|
| 146 |
-
"116": "S-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 147 |
},
|
| 148 |
"initializer_range": 0.02,
|
| 149 |
"intermediate_size": 6656,
|
|
@@ -155,28 +167,31 @@
|
|
| 155 |
"B-contact.postal_code": 17,
|
| 156 |
"B-credential.api_key": 21,
|
| 157 |
"B-credential.password": 25,
|
| 158 |
-
"B-
|
| 159 |
-
"B-
|
| 160 |
-
"B-
|
| 161 |
-
"B-financial.
|
| 162 |
-
"B-financial.
|
| 163 |
-
"B-
|
| 164 |
-
"B-
|
| 165 |
-
"B-healthcare.
|
| 166 |
-
"B-
|
| 167 |
-
"B-
|
| 168 |
-
"B-
|
| 169 |
-
"B-identity.
|
| 170 |
-
"B-identity.
|
| 171 |
-
"B-identity.
|
| 172 |
-
"B-identity.
|
| 173 |
-
"B-
|
| 174 |
-
"B-
|
| 175 |
-
"B-
|
| 176 |
-
"B-
|
| 177 |
-
"B-
|
| 178 |
-
"B-
|
| 179 |
-
"B-
|
|
|
|
|
|
|
|
|
|
| 180 |
"E-contact.address": 3,
|
| 181 |
"E-contact.email": 7,
|
| 182 |
"E-contact.ip_address": 11,
|
|
@@ -184,28 +199,31 @@
|
|
| 184 |
"E-contact.postal_code": 19,
|
| 185 |
"E-credential.api_key": 23,
|
| 186 |
"E-credential.password": 27,
|
| 187 |
-
"E-
|
| 188 |
-
"E-
|
| 189 |
-
"E-
|
| 190 |
-
"E-financial.
|
| 191 |
-
"E-financial.
|
| 192 |
-
"E-
|
| 193 |
-
"E-
|
| 194 |
-
"E-healthcare.
|
| 195 |
-
"E-
|
| 196 |
-
"E-
|
| 197 |
-
"E-
|
| 198 |
-
"E-identity.
|
| 199 |
-
"E-identity.
|
| 200 |
-
"E-identity.
|
| 201 |
-
"E-identity.
|
| 202 |
-
"E-
|
| 203 |
-
"E-
|
| 204 |
-
"E-
|
| 205 |
-
"E-
|
| 206 |
-
"E-
|
| 207 |
-
"E-
|
| 208 |
-
"E-
|
|
|
|
|
|
|
|
|
|
| 209 |
"I-contact.address": 2,
|
| 210 |
"I-contact.email": 6,
|
| 211 |
"I-contact.ip_address": 10,
|
|
@@ -213,28 +231,31 @@
|
|
| 213 |
"I-contact.postal_code": 18,
|
| 214 |
"I-credential.api_key": 22,
|
| 215 |
"I-credential.password": 26,
|
| 216 |
-
"I-
|
| 217 |
-
"I-
|
| 218 |
-
"I-
|
| 219 |
-
"I-financial.
|
| 220 |
-
"I-financial.
|
| 221 |
-
"I-
|
| 222 |
-
"I-
|
| 223 |
-
"I-healthcare.
|
| 224 |
-
"I-
|
| 225 |
-
"I-
|
| 226 |
-
"I-
|
| 227 |
-
"I-identity.
|
| 228 |
-
"I-identity.
|
| 229 |
-
"I-identity.
|
| 230 |
-
"I-identity.
|
| 231 |
-
"I-
|
| 232 |
-
"I-
|
| 233 |
-
"I-
|
| 234 |
-
"I-
|
| 235 |
-
"I-
|
| 236 |
-
"I-
|
| 237 |
-
"I-
|
|
|
|
|
|
|
|
|
|
| 238 |
"O": 0,
|
| 239 |
"S-contact.address": 4,
|
| 240 |
"S-contact.email": 8,
|
|
@@ -243,28 +264,31 @@
|
|
| 243 |
"S-contact.postal_code": 20,
|
| 244 |
"S-credential.api_key": 24,
|
| 245 |
"S-credential.password": 28,
|
| 246 |
-
"S-
|
| 247 |
-
"S-
|
| 248 |
-
"S-
|
| 249 |
-
"S-financial.
|
| 250 |
-
"S-financial.
|
| 251 |
-
"S-
|
| 252 |
-
"S-
|
| 253 |
-
"S-healthcare.
|
| 254 |
-
"S-
|
| 255 |
-
"S-
|
| 256 |
-
"S-
|
| 257 |
-
"S-identity.
|
| 258 |
-
"S-identity.
|
| 259 |
-
"S-identity.
|
| 260 |
-
"S-identity.
|
| 261 |
-
"S-
|
| 262 |
-
"S-
|
| 263 |
-
"S-
|
| 264 |
-
"S-
|
| 265 |
-
"S-
|
| 266 |
-
"S-
|
| 267 |
-
"S-
|
|
|
|
|
|
|
|
|
|
| 268 |
},
|
| 269 |
"label_smoothing": 0.1,
|
| 270 |
"layer_types": [
|
|
|
|
| 56 |
"26": "I-credential.password",
|
| 57 |
"27": "E-credential.password",
|
| 58 |
"28": "S-credential.password",
|
| 59 |
+
"29": "B-developer.device_id",
|
| 60 |
+
"30": "I-developer.device_id",
|
| 61 |
+
"31": "E-developer.device_id",
|
| 62 |
+
"32": "S-developer.device_id",
|
| 63 |
+
"33": "B-developer.login_credentials",
|
| 64 |
+
"34": "I-developer.login_credentials",
|
| 65 |
+
"35": "E-developer.login_credentials",
|
| 66 |
+
"36": "S-developer.login_credentials",
|
| 67 |
+
"37": "B-device.mac_address",
|
| 68 |
+
"38": "I-device.mac_address",
|
| 69 |
+
"39": "E-device.mac_address",
|
| 70 |
+
"40": "S-device.mac_address",
|
| 71 |
+
"41": "B-financial.bank_account",
|
| 72 |
+
"42": "I-financial.bank_account",
|
| 73 |
+
"43": "E-financial.bank_account",
|
| 74 |
+
"44": "S-financial.bank_account",
|
| 75 |
+
"45": "B-financial.credit_card",
|
| 76 |
+
"46": "I-financial.credit_card",
|
| 77 |
+
"47": "E-financial.credit_card",
|
| 78 |
+
"48": "S-financial.credit_card",
|
| 79 |
+
"49": "B-financial.iban",
|
| 80 |
+
"50": "I-financial.iban",
|
| 81 |
+
"51": "E-financial.iban",
|
| 82 |
+
"52": "S-financial.iban",
|
| 83 |
+
"53": "B-financial.swift_bic",
|
| 84 |
+
"54": "I-financial.swift_bic",
|
| 85 |
+
"55": "E-financial.swift_bic",
|
| 86 |
+
"56": "S-financial.swift_bic",
|
| 87 |
+
"57": "B-healthcare.condition",
|
| 88 |
+
"58": "I-healthcare.condition",
|
| 89 |
+
"59": "E-healthcare.condition",
|
| 90 |
+
"60": "S-healthcare.condition",
|
| 91 |
+
"61": "B-healthcare.health_plan_id",
|
| 92 |
+
"62": "I-healthcare.health_plan_id",
|
| 93 |
+
"63": "E-healthcare.health_plan_id",
|
| 94 |
+
"64": "S-healthcare.health_plan_id",
|
| 95 |
+
"65": "B-healthcare.medical_record",
|
| 96 |
+
"66": "I-healthcare.medical_record",
|
| 97 |
+
"67": "E-healthcare.medical_record",
|
| 98 |
+
"68": "S-healthcare.medical_record",
|
| 99 |
+
"69": "B-healthcare.medication",
|
| 100 |
+
"70": "I-healthcare.medication",
|
| 101 |
+
"71": "E-healthcare.medication",
|
| 102 |
+
"72": "S-healthcare.medication",
|
| 103 |
+
"73": "B-identity.date_of_birth",
|
| 104 |
+
"74": "I-identity.date_of_birth",
|
| 105 |
+
"75": "E-identity.date_of_birth",
|
| 106 |
+
"76": "S-identity.date_of_birth",
|
| 107 |
+
"77": "B-identity.drivers_license",
|
| 108 |
+
"78": "I-identity.drivers_license",
|
| 109 |
+
"79": "E-identity.drivers_license",
|
| 110 |
+
"80": "S-identity.drivers_license",
|
| 111 |
+
"81": "B-identity.national_id",
|
| 112 |
+
"82": "I-identity.national_id",
|
| 113 |
+
"83": "E-identity.national_id",
|
| 114 |
+
"84": "S-identity.national_id",
|
| 115 |
+
"85": "B-identity.passport",
|
| 116 |
+
"86": "I-identity.passport",
|
| 117 |
+
"87": "E-identity.passport",
|
| 118 |
+
"88": "S-identity.passport",
|
| 119 |
+
"89": "B-identity.person_name",
|
| 120 |
+
"90": "I-identity.person_name",
|
| 121 |
+
"91": "E-identity.person_name",
|
| 122 |
+
"92": "S-identity.person_name",
|
| 123 |
+
"93": "B-identity.ssn",
|
| 124 |
+
"94": "I-identity.ssn",
|
| 125 |
+
"95": "E-identity.ssn",
|
| 126 |
+
"96": "S-identity.ssn",
|
| 127 |
+
"97": "B-identity.tax_id",
|
| 128 |
+
"98": "I-identity.tax_id",
|
| 129 |
+
"99": "E-identity.tax_id",
|
| 130 |
+
"100": "S-identity.tax_id",
|
| 131 |
+
"101": "B-legal.case_number",
|
| 132 |
+
"102": "I-legal.case_number",
|
| 133 |
+
"103": "E-legal.case_number",
|
| 134 |
+
"104": "S-legal.case_number",
|
| 135 |
+
"105": "B-location.gps_coordinates",
|
| 136 |
+
"106": "I-location.gps_coordinates",
|
| 137 |
+
"107": "E-location.gps_coordinates",
|
| 138 |
+
"108": "S-location.gps_coordinates",
|
| 139 |
+
"109": "B-online.url",
|
| 140 |
+
"110": "I-online.url",
|
| 141 |
+
"111": "E-online.url",
|
| 142 |
+
"112": "S-online.url",
|
| 143 |
+
"113": "B-online.username",
|
| 144 |
+
"114": "I-online.username",
|
| 145 |
+
"115": "E-online.username",
|
| 146 |
+
"116": "S-online.username",
|
| 147 |
+
"117": "B-org.company_name",
|
| 148 |
+
"118": "I-org.company_name",
|
| 149 |
+
"119": "E-org.company_name",
|
| 150 |
+
"120": "S-org.company_name",
|
| 151 |
+
"121": "B-special.political",
|
| 152 |
+
"122": "I-special.political",
|
| 153 |
+
"123": "E-special.political",
|
| 154 |
+
"124": "S-special.political",
|
| 155 |
+
"125": "B-special.religion",
|
| 156 |
+
"126": "I-special.religion",
|
| 157 |
+
"127": "E-special.religion",
|
| 158 |
+
"128": "S-special.religion"
|
| 159 |
},
|
| 160 |
"initializer_range": 0.02,
|
| 161 |
"intermediate_size": 6656,
|
|
|
|
| 167 |
"B-contact.postal_code": 17,
|
| 168 |
"B-credential.api_key": 21,
|
| 169 |
"B-credential.password": 25,
|
| 170 |
+
"B-developer.device_id": 29,
|
| 171 |
+
"B-developer.login_credentials": 33,
|
| 172 |
+
"B-device.mac_address": 37,
|
| 173 |
+
"B-financial.bank_account": 41,
|
| 174 |
+
"B-financial.credit_card": 45,
|
| 175 |
+
"B-financial.iban": 49,
|
| 176 |
+
"B-financial.swift_bic": 53,
|
| 177 |
+
"B-healthcare.condition": 57,
|
| 178 |
+
"B-healthcare.health_plan_id": 61,
|
| 179 |
+
"B-healthcare.medical_record": 65,
|
| 180 |
+
"B-healthcare.medication": 69,
|
| 181 |
+
"B-identity.date_of_birth": 73,
|
| 182 |
+
"B-identity.drivers_license": 77,
|
| 183 |
+
"B-identity.national_id": 81,
|
| 184 |
+
"B-identity.passport": 85,
|
| 185 |
+
"B-identity.person_name": 89,
|
| 186 |
+
"B-identity.ssn": 93,
|
| 187 |
+
"B-identity.tax_id": 97,
|
| 188 |
+
"B-legal.case_number": 101,
|
| 189 |
+
"B-location.gps_coordinates": 105,
|
| 190 |
+
"B-online.url": 109,
|
| 191 |
+
"B-online.username": 113,
|
| 192 |
+
"B-org.company_name": 117,
|
| 193 |
+
"B-special.political": 121,
|
| 194 |
+
"B-special.religion": 125,
|
| 195 |
"E-contact.address": 3,
|
| 196 |
"E-contact.email": 7,
|
| 197 |
"E-contact.ip_address": 11,
|
|
|
|
| 199 |
"E-contact.postal_code": 19,
|
| 200 |
"E-credential.api_key": 23,
|
| 201 |
"E-credential.password": 27,
|
| 202 |
+
"E-developer.device_id": 31,
|
| 203 |
+
"E-developer.login_credentials": 35,
|
| 204 |
+
"E-device.mac_address": 39,
|
| 205 |
+
"E-financial.bank_account": 43,
|
| 206 |
+
"E-financial.credit_card": 47,
|
| 207 |
+
"E-financial.iban": 51,
|
| 208 |
+
"E-financial.swift_bic": 55,
|
| 209 |
+
"E-healthcare.condition": 59,
|
| 210 |
+
"E-healthcare.health_plan_id": 63,
|
| 211 |
+
"E-healthcare.medical_record": 67,
|
| 212 |
+
"E-healthcare.medication": 71,
|
| 213 |
+
"E-identity.date_of_birth": 75,
|
| 214 |
+
"E-identity.drivers_license": 79,
|
| 215 |
+
"E-identity.national_id": 83,
|
| 216 |
+
"E-identity.passport": 87,
|
| 217 |
+
"E-identity.person_name": 91,
|
| 218 |
+
"E-identity.ssn": 95,
|
| 219 |
+
"E-identity.tax_id": 99,
|
| 220 |
+
"E-legal.case_number": 103,
|
| 221 |
+
"E-location.gps_coordinates": 107,
|
| 222 |
+
"E-online.url": 111,
|
| 223 |
+
"E-online.username": 115,
|
| 224 |
+
"E-org.company_name": 119,
|
| 225 |
+
"E-special.political": 123,
|
| 226 |
+
"E-special.religion": 127,
|
| 227 |
"I-contact.address": 2,
|
| 228 |
"I-contact.email": 6,
|
| 229 |
"I-contact.ip_address": 10,
|
|
|
|
| 231 |
"I-contact.postal_code": 18,
|
| 232 |
"I-credential.api_key": 22,
|
| 233 |
"I-credential.password": 26,
|
| 234 |
+
"I-developer.device_id": 30,
|
| 235 |
+
"I-developer.login_credentials": 34,
|
| 236 |
+
"I-device.mac_address": 38,
|
| 237 |
+
"I-financial.bank_account": 42,
|
| 238 |
+
"I-financial.credit_card": 46,
|
| 239 |
+
"I-financial.iban": 50,
|
| 240 |
+
"I-financial.swift_bic": 54,
|
| 241 |
+
"I-healthcare.condition": 58,
|
| 242 |
+
"I-healthcare.health_plan_id": 62,
|
| 243 |
+
"I-healthcare.medical_record": 66,
|
| 244 |
+
"I-healthcare.medication": 70,
|
| 245 |
+
"I-identity.date_of_birth": 74,
|
| 246 |
+
"I-identity.drivers_license": 78,
|
| 247 |
+
"I-identity.national_id": 82,
|
| 248 |
+
"I-identity.passport": 86,
|
| 249 |
+
"I-identity.person_name": 90,
|
| 250 |
+
"I-identity.ssn": 94,
|
| 251 |
+
"I-identity.tax_id": 98,
|
| 252 |
+
"I-legal.case_number": 102,
|
| 253 |
+
"I-location.gps_coordinates": 106,
|
| 254 |
+
"I-online.url": 110,
|
| 255 |
+
"I-online.username": 114,
|
| 256 |
+
"I-org.company_name": 118,
|
| 257 |
+
"I-special.political": 122,
|
| 258 |
+
"I-special.religion": 126,
|
| 259 |
"O": 0,
|
| 260 |
"S-contact.address": 4,
|
| 261 |
"S-contact.email": 8,
|
|
|
|
| 264 |
"S-contact.postal_code": 20,
|
| 265 |
"S-credential.api_key": 24,
|
| 266 |
"S-credential.password": 28,
|
| 267 |
+
"S-developer.device_id": 32,
|
| 268 |
+
"S-developer.login_credentials": 36,
|
| 269 |
+
"S-device.mac_address": 40,
|
| 270 |
+
"S-financial.bank_account": 44,
|
| 271 |
+
"S-financial.credit_card": 48,
|
| 272 |
+
"S-financial.iban": 52,
|
| 273 |
+
"S-financial.swift_bic": 56,
|
| 274 |
+
"S-healthcare.condition": 60,
|
| 275 |
+
"S-healthcare.health_plan_id": 64,
|
| 276 |
+
"S-healthcare.medical_record": 68,
|
| 277 |
+
"S-healthcare.medication": 72,
|
| 278 |
+
"S-identity.date_of_birth": 76,
|
| 279 |
+
"S-identity.drivers_license": 80,
|
| 280 |
+
"S-identity.national_id": 84,
|
| 281 |
+
"S-identity.passport": 88,
|
| 282 |
+
"S-identity.person_name": 92,
|
| 283 |
+
"S-identity.ssn": 96,
|
| 284 |
+
"S-identity.tax_id": 100,
|
| 285 |
+
"S-legal.case_number": 104,
|
| 286 |
+
"S-location.gps_coordinates": 108,
|
| 287 |
+
"S-online.url": 112,
|
| 288 |
+
"S-online.username": 116,
|
| 289 |
+
"S-org.company_name": 120,
|
| 290 |
+
"S-special.political": 124,
|
| 291 |
+
"S-special.religion": 128
|
| 292 |
},
|
| 293 |
"label_smoothing": 0.1,
|
| 294 |
"layer_types": [
|
leaderboard.png
CHANGED
|
Git LFS Details
|
|
Git LFS Details
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a8e525b1856af936e8feb1b35fa1fb25e5f39bfaf78b7319f0f39fe8bfb15e58
|
| 3 |
+
size 1250706876
|