Release: masker-mini

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

LICENSE

@@ -0,0 +1,91 @@
+# Offchain Studio Source License, Version 1.0
+
+Dated: 6 July 2026
+
+Licensor: Offchain Studio, registered with the Dutch Chamber of Commerce (Kamer van Koophandel) under KvK number 92717810.
+
+This is a source-available license. It is not an open source license as defined by the Open Source Initiative, because it limits free use by size.
+
+## 1. Acceptance
+
+By using the Software you agree to this license. If you do not agree, you have no license. The obligations in this license are both conditions of the licenses granted to you and independent obligations you owe the Licensor.
+
+## 2. Definitions
+
+- **Licensor** means Offchain Studio, KvK number 92717810.
+- **Software** means each work that the Licensor releases under this license, including its source code, model weights, and documentation, as identified in the notice that accompanies that work.
+- **Notice** means the file the Licensor distributes with the Software that identifies the Software, its functional category, the Licensor, and the contact address for commercial licensing.
+- **Output** means any data the Software produces, including predictions, labels, redactions, embeddings, logits, and generated text.
+- **You** means the individual or legal entity exercising rights under this license.
+- **Your Company** means you, together with every entity that controls you, is controlled by you, or is under common control with you. Control means holding a majority of voting rights, or the power to direct management by contract or otherwise.
+- **Use** means anything you do with the Software that requires one of the licenses granted here.
+- **Monthly Active Devices** (**MAD**) means the number of distinct end-user devices on which the Software executes at least once during a calendar month, counted across all applications and services of Your Company.
+- **Revenue** means the total worldwide gross revenue of Your Company over its most recent completed financial year, determined under the accounting standards Your Company applies to its statutory accounts.
+- **Permitted Use** has the meaning given in Section 4.
+- **Commercial License** means a separate written license from the Licensor covering use beyond Permitted Use.
+
+## 3. Copyright and Patent License
+
+Subject to this license, the Licensor grants you a worldwide, royalty-free, non-exclusive license to use, reproduce, modify, create derivative works of, and distribute the Software, solely for Permitted Use.
+
+The Licensor also grants you, for Permitted Use, a license under patent claims the Licensor can license that you would infringe by using the Software as provided. This patent license ends immediately if you or Your Company assert, in any forum, that the Software infringes a patent.
+
+## 4. Permitted Use (free tier)
+
+Your Use is Permitted Use only while **both** of the following are true for Your Company:
+
+(a) MAD does not exceed **150,000** in any calendar month; and
+
+(b) Revenue does not exceed **EUR 1,000,000**.
+
+If **either** threshold is exceeded, your Use stops being Permitted Use, and you need a Commercial License to continue using the Software.
+
+## 5. Commercial License
+
+If your Use is not Permitted Use, you must obtain a Commercial License before you continue using the Software.
+
+The Licensor will offer a Commercial License on fair and reasonable terms, and will not discriminate between licensees in comparable situations. Request one at the contact address stated in the Notice.
+
+## 6. Restrictions on training other models
+
+Section 3 lets you modify the Software and create derivative works of it for Permitted Use. This section limits how you may use the Software and its Output to build a separate model.
+
+You may not use the Software, its Output, or logs of its Output to train, fine-tune, distill, or otherwise develop a machine learning model that provides functionality substantially similar to the Software. Where the Notice states a functional category for the Software, a model that provides functionality in that category is treated as substantially similar. Creating a derivative work of the Software itself under Section 3 is not restricted by this section.
+
+This restriction is a contractual obligation and does not depend on whether copyright protects the Output. The Licensor expressly reserves, under Article 4(3) of Directive (EU) 2019/790, the right to opt out of text and data mining of the Software and its Output.
+
+## 7. Conditions on distribution
+
+When you distribute the Software or a derivative of it, you must:
+
+(a) include a copy of this license, or its URL, and a copy of the Notice; and
+
+(b) keep intact all copyright and license notices, and any line beginning "Required Notice:" that the Licensor includes with the Software.
+
+You may not sublicense the Software. Each recipient receives its license directly from the Licensor under these terms.
+
+## 8. Verification
+
+Once per calendar year, on 14 days' written notice, the Licensor may ask you to confirm in writing Your Company's MAD and Revenue for the relevant period. You must respond accurately. The Licensor will keep that information confidential and use it only to check compliance with this license.
+
+## 9. Term and termination
+
+This license lasts as long as the Licensor holds copyright in the Software.
+
+If you breach this license, your licenses end automatically. If the Licensor notifies you of the breach in writing, and you cure it and take reasonable steps to correct past breaches within 30 days of that notice, your licenses are reinstated. This reinstatement applies only to your first such notice. A later breach ends your licenses permanently.
+
+## 10. No warranty
+
+To the maximum extent permitted by law, the Software is provided "as is", without any warranty or condition. The Licensor does not warrant that the Software is error-free, accurate, or fit for any particular purpose.
+
+## 11. Liability
+
+To the maximum extent permitted by law, the Licensor is not liable for any damage arising out of this license or out of use of the Software. This limitation does not apply to damage caused by the Licensor's intent (opzet) or deliberate recklessness (bewuste roekeloosheid), and does not exclude any liability that cannot be excluded under mandatory Dutch law.
+
+## 12. Governing law and jurisdiction
+
+This license is governed by the laws of the Netherlands, excluding its conflict-of-law rules and the UN Convention on Contracts for the International Sale of Goods (CISG). Disputes arising out of or relating to this license are submitted exclusively to the District Court of Amsterdam (Rechtbank Amsterdam), without prejudice to any jurisdiction that mandatory law assigns elsewhere, including mandatory consumer rules.
+
+## 13. Miscellaneous
+
+If any provision of this license is held unenforceable, the rest stays in effect, and the unenforceable provision applies to the maximum extent permitted. For Permitted Use, this license is the entire agreement about the Software between you and the Licensor. Only the Licensor can change this license, by publishing a new version.

NOTICE

@@ -0,0 +1,9 @@
+This work is licensed under the Offchain Studio Source License, Version 1.0.
+Full terms: https://ai.basement.dev/license
+
+Software: MASKER-MINI
+Functional category: PII detection and redaction
+Licensor: Offchain Studio, KvK 92717810
+Commercial license requests: licensing@basement.dev
+
+Required Notice: MASKER-MINI (c) 2026 Offchain Studio

README.md

@@ -0,0 +1,216 @@
+---
+license: other
+license_name: offchain-studio-source-license-1.0
+license_link: https://ai.basement.dev/license
+library_name: transformers
+pipeline_tag: token-classification
+base_model: amsintelligence/masker
+base_model_relation: finetune
+datasets:
+  - ai4privacy/pii-masking-openpii-1m
+language:
+  - bg
+  - cs
+  - da
+  - de
+  - el
+  - en
+  - es
+  - et
+  - fi
+  - fr
+  - hr
+  - hu
+  - it
+  - lt
+  - lv
+  - nl
+  - pl
+  - pt
+  - ro
+  - sk
+  - sl
+  - sr
+  - sv
+tags:
+  - token-classification
+  - pii
+  - pii-detection
+  - pii-masking
+  - redaction
+  - privacy
+  - on-device
+  - onnx
+  - coreml
+  - distillation
+metrics:
+  - f1
+model-index:
+  - name: masker-mini
+    results:
+      - task:
+          type: token-classification
+          name: PII span detection
+        dataset:
+          name: pii-masking-openpii-1m (validation)
+          type: ai4privacy/pii-masking-openpii-1m
+        metrics:
+          - type: f1
+            name: Strict span F1
+            value: 0.965
+          - type: f1
+            name: Typed F1
+            value: 0.991
+          - type: recall
+            name: Leak-safe recall
+            value: 0.998
+---
+
+# Masker Mini
+
+**Masker-mini** is the on-device sibling of
+[**masker**](https://huggingface.co/amsintelligence/masker): a 6-layer distilled
+PII detector for **23 European languages** that keeps ~99% of the teacher's
+strict-span F1 while shrinking to as little as **17 MB** (4-bit Core ML).
+
+It runs fully offline, no PII leaves the device.
+
+| artifact                            | format                   | size      | strict F1 |
+| ----------------------------------- | ------------------------ | --------- | --------- |
+| `model.safetensors`                 | PyTorch fp16             | 68 MB     | 0.965     |
+| `onnx/model_fp16.onnx`              | ONNX fp16 (portable)      | 68 MB     | 0.965     |
+| `onnx/model_int4.onnx`              | ONNX 4-bit (ONNX Runtime) | **22 MB** | 0.965     |
+| `coreml/masker_mini_4bit.mlpackage` | Core ML 4-bit (Apple NE)  | **17 MB** | 0.965     |
+
+## Model type & training
+
+Masker Mini is a **6-layer BERT-architecture token classifier** (hidden size 384,
+12 heads, ~35.6M parameters), a MiniLM-class encoder. It is trained by **knowledge
+distillation** from `masker`.
+
+After distillation the vocabulary was frequency-pruned,
+shrinking the embedding table to land the whole network at ~35.6M parameters.
+
+Three deployment artifacts are provided:
+
+- **ONNX fp16** (68 MB): portable, runs anywhere via ONNX Runtime
+  (Android / iOS / web / server); numerically identical to the PyTorch model.
+- **ONNX 4-bit** (22 MB): weight-only block quantization (`MatMulNBits` +
+  `GatherBlockQuantized`) of the Linear layers _and_ the embedding table. Needs
+  ONNX Runtime ≥ 1.18 for the 4-bit ops; 99.8% token-faithful to fp16.
+- **Core ML, 4-bit palettized** (17 MB): weight-only k-means palettization for
+  the Apple Neural Engine (iOS 18 / macOS 15+). Palettization is effectively
+  lossless here (Δ strict F1 = −0.0002 vs fp16).
+
+It emits the **same 12 entity types** as
+[masker](https://huggingface.co/amsintelligence/masker) (48 BIOES labels + `O`)
+and slots into the same rules-layer pipeline for structured PII.
+
+## Usage
+
+For a plain PyTorch / Transformers quick start, the snippet on the
+[masker](https://huggingface.co/amsintelligence/masker) card runs unchanged, just
+point it at `amsintelligence/masker-mini`.
+
+What this repo is actually for is the two on-device builds:
+
+**ONNX Runtime** (portable for Android, iOS, web, server)
+
+```python
+import onnxruntime as ort, numpy as np
+from transformers import AutoTokenizer
+
+tok = AutoTokenizer.from_pretrained("amsintelligence/masker-mini")
+sess = ort.InferenceSession("onnx/model_fp16.onnx")
+text = "Stuur de factuur naar Sanne de Groot in Utrecht."
+feed = {k: v.astype(np.int64) for k, v in tok(text, return_tensors="np").items()}
+logits = sess.run(None, feed)[0]        # [1, seq_len, 49] BIOES logits -> argmax + decode
+```
+
+For the smallest portable build, swap in `onnx/model_int4.onnx` (22 MB, 4-bit) —
+same inputs/outputs, needs ONNX Runtime ≥ 1.18.
+
+**Core ML**, add `coreml/masker_mini_4bit.mlpackage` to an Xcode target.
+
+Inputs `input_ids`, `attention_mask`, `token_type_ids`
+(`Int32`, fixed length 256); output `logits`. The BIOES→span decode is the same
+handful of lines shown on the masker card.
+
+## Evaluation
+
+openpii-1m validation, span-level, boundary-exact. **In-distribution — see
+Limitations.** Numbers below are the **4-bit Core ML** build (fp16/PyTorch are
+within ±0.001).
+
+**Overall:** strict F1 **0.965** · typed F1 0.991 · leak-safe recall 0.998.
+Distillation gap vs the masker teacher (0.982): **−0.007**.
+
+**Per-type strict F1**
+
+| entity        | F1    | entity          | F1    |
+| ------------- | ----- | --------------- | ----- |
+| DATE          | 0.999 | AGE             | 0.969 |
+| EMAIL         | 0.999 | BUILDING_NUMBER | 0.969 |
+| CREDIT_CARD   | 0.999 | CITY            | 0.954 |
+| PHONE         | 0.998 | STREET_NAME     | 0.922 |
+| GOVERNMENT_ID | 0.998 | GIVEN_NAME      | 0.910 |
+| ZIP_CODE      | 0.994 | SURNAME         | 0.900 |
+
+**By language**
+
+| split                          | strict F1     |
+| ------------------------------ | ------------- |
+| English                        | 0.974         |
+| Non-English (22 langs, pooled) | 0.971         |
+| Dutch (flagship)               | 0.968         |
+| Per-language range             | 0.960 - 0.983 |
+
+## Limitations & biases
+
+- **This is a compressed model.** It trails the full-size masker by ~0.7 strict-F1
+  points, and the loss is not uniform: it lands almost entirely on `GIVEN_NAME` /
+  `SURNAME`, plus `CITY` / `STREET_NAME` from the reduced 64K vocabulary. Structured
+  types (email, phone, IDs, cards, dates) stay ≥ 0.99. The one language-specific
+  soft spot is Dutch person-name boundaries (_tussenvoegsels_). If you need the top
+  half-point back, use masker.
+- Scores are **in-distribution** (synthetic openpii), treat them as an upper
+  bound, measure on your own text, and design for residual leakage rather than
+  assuming full coverage.
+
+## Credits & attribution
+
+Distilled from **masker**, which is itself a derivative of:
+
+- **Backbone / tokenizer lineage:** [`microsoft/mdeberta-v3-base`](https://huggingface.co/microsoft/mdeberta-v3-base)
+  by Microsoft; **MIT License** (masker-mini reuses mDeBERTa's SentencePiece
+  tokenizer). DeBERTaV3: He, Gao & Chen, 2021
+  ([arXiv:2111.09543](https://arxiv.org/abs/2111.09543)).
+- **Training data:** [`ai4privacy/pii-masking-openpii-1m`](https://huggingface.co/datasets/ai4privacy/pii-masking-openpii-1m)
+  by **Ai4Privacy**; **CC-BY-4.0**. Attribution required; please retain this
+  credit in downstream use.
+
+## License
+
+Licensed under the **Offchain Studio Source License, Version 1.0**, a
+source-available license. Full terms: <https://ai.basement.dev/license>.
+
+Commercial-license requests: `licensing@basement.dev`.
+
+See the accompanying **`NOTICE`** file; its Required
+Notice line _MASKER-MINI © 2026 Offchain Studio_ must be retained in
+redistribution.
+
+The underlying components keep their own (attribution-only) terms, retained in
+Credits above: the mDeBERTa lineage is **MIT** and the openpii training data is
+**CC-BY-4.0**.
+
+## Citation
+
+```bibtex
+@software{masker_mini,
+  title  = {masker-mini: on-device PII detection for 23 European languages},
+  year   = {2026},
+  note   = {6-layer distillation of masker (mDeBERTa-v3); ONNX + Core ML},
+  url    = {https://huggingface.co/amsintelligence/masker-mini}
+}
+```

added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "[MASK]": 64360
+}

config.json

@@ -0,0 +1,126 @@
+{
+  "architectures": [
+    "BertForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "dtype": "float16",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 384,
+  "id2label": {
+    "0": "O",
+    "1": "B-AGE",
+    "2": "I-AGE",
+    "3": "E-AGE",
+    "4": "S-AGE",
+    "5": "B-BUILDING_NUMBER",
+    "6": "I-BUILDING_NUMBER",
+    "7": "E-BUILDING_NUMBER",
+    "8": "S-BUILDING_NUMBER",
+    "9": "B-CITY",
+    "10": "I-CITY",
+    "11": "E-CITY",
+    "12": "S-CITY",
+    "13": "B-CREDIT_CARD",
+    "14": "I-CREDIT_CARD",
+    "15": "E-CREDIT_CARD",
+    "16": "S-CREDIT_CARD",
+    "17": "B-DATE",
+    "18": "I-DATE",
+    "19": "E-DATE",
+    "20": "S-DATE",
+    "21": "B-EMAIL",
+    "22": "I-EMAIL",
+    "23": "E-EMAIL",
+    "24": "S-EMAIL",
+    "25": "B-GIVEN_NAME",
+    "26": "I-GIVEN_NAME",
+    "27": "E-GIVEN_NAME",
+    "28": "S-GIVEN_NAME",
+    "29": "B-GOVERNMENT_ID",
+    "30": "I-GOVERNMENT_ID",
+    "31": "E-GOVERNMENT_ID",
+    "32": "S-GOVERNMENT_ID",
+    "33": "B-PHONE",
+    "34": "I-PHONE",
+    "35": "E-PHONE",
+    "36": "S-PHONE",
+    "37": "B-STREET_NAME",
+    "38": "I-STREET_NAME",
+    "39": "E-STREET_NAME",
+    "40": "S-STREET_NAME",
+    "41": "B-SURNAME",
+    "42": "I-SURNAME",
+    "43": "E-SURNAME",
+    "44": "S-SURNAME",
+    "45": "B-ZIP_CODE",
+    "46": "I-ZIP_CODE",
+    "47": "E-ZIP_CODE",
+    "48": "S-ZIP_CODE"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 1536,
+  "label2id": {
+    "B-AGE": 1,
+    "B-BUILDING_NUMBER": 5,
+    "B-CITY": 9,
+    "B-CREDIT_CARD": 13,
+    "B-DATE": 17,
+    "B-EMAIL": 21,
+    "B-GIVEN_NAME": 25,
+    "B-GOVERNMENT_ID": 29,
+    "B-PHONE": 33,
+    "B-STREET_NAME": 37,
+    "B-SURNAME": 41,
+    "B-ZIP_CODE": 45,
+    "E-AGE": 3,
+    "E-BUILDING_NUMBER": 7,
+    "E-CITY": 11,
+    "E-CREDIT_CARD": 15,
+    "E-DATE": 19,
+    "E-EMAIL": 23,
+    "E-GIVEN_NAME": 27,
+    "E-GOVERNMENT_ID": 31,
+    "E-PHONE": 35,
+    "E-STREET_NAME": 39,
+    "E-SURNAME": 43,
+    "E-ZIP_CODE": 47,
+    "I-AGE": 2,
+    "I-BUILDING_NUMBER": 6,
+    "I-CITY": 10,
+    "I-CREDIT_CARD": 14,
+    "I-DATE": 18,
+    "I-EMAIL": 22,
+    "I-GIVEN_NAME": 26,
+    "I-GOVERNMENT_ID": 30,
+    "I-PHONE": 34,
+    "I-STREET_NAME": 38,
+    "I-SURNAME": 42,
+    "I-ZIP_CODE": 46,
+    "O": 0,
+    "S-AGE": 4,
+    "S-BUILDING_NUMBER": 8,
+    "S-CITY": 12,
+    "S-CREDIT_CARD": 16,
+    "S-DATE": 20,
+    "S-EMAIL": 24,
+    "S-GIVEN_NAME": 28,
+    "S-GOVERNMENT_ID": 32,
+    "S-PHONE": 36,
+    "S-STREET_NAME": 40,
+    "S-SURNAME": 44,
+    "S-ZIP_CODE": 48
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.57.6",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 64361
+}

coreml/masker_mini_4bit.mlpackage/Data/com.apple.CoreML/model.mlmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:916c7e5e84375ad6fcb0a5be73a1f8d3abefc14a3dd5ebad0cb76650c90a738a
+size 71875

coreml/masker_mini_4bit.mlpackage/Data/com.apple.CoreML/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49a3b4d2f75fd58bba47dd4183f8ca71a937379a8df81786cb0438b86883462d
+size 17799010

coreml/masker_mini_4bit.mlpackage/Manifest.json

@@ -0,0 +1,18 @@
+{
+    "fileFormatVersion": "1.0.0",
+    "itemInfoEntries": {
+        "67CFA8D6-3E3D-4E47-A023-ACA05483E764": {
+            "author": "com.apple.CoreML",
+            "description": "CoreML Model Weights",
+            "name": "weights",
+            "path": "com.apple.CoreML/weights"
+        },
+        "7C701565-922A-4EEA-A9DF-1F68448CF5EB": {
+            "author": "com.apple.CoreML",
+            "description": "CoreML Model Specification",
+            "name": "model.mlmodel",
+            "path": "com.apple.CoreML/model.mlmodel"
+        }
+    },
+    "rootModelIdentifier": "7C701565-922A-4EEA-A9DF-1F68448CF5EB"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75b05de44ea87720a5226fc8a688663df99cb4601e6a08823c2728628d1de2b2
+size 71168658

onnx/model_fp16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bcece622496c55835ac485bda83b5ae9c55cf070bceb8ba3fcf7e7e20f4a9652
+size 71308586

onnx/model_int4.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed732502da5c5dcaaa3c8b07b9af0c295a3440d66fd39b82ffe12b055581c712
+size 22456156

special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e722c36bad602a200b3d9dbba7684150edb5449bf3b1494839030107004b09b
+size 1215690

tokenizer_config.json

@@ -0,0 +1,858 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "64260": {
+      "content": "▁<extra_id_99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64261": {
+      "content": "▁<extra_id_98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64262": {
+      "content": "▁<extra_id_97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64263": {
+      "content": "▁<extra_id_96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64264": {
+      "content": "▁<extra_id_95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64265": {
+      "content": "▁<extra_id_94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64266": {
+      "content": "▁<extra_id_93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64267": {
+      "content": "▁<extra_id_92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64268": {
+      "content": "▁<extra_id_91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64269": {
+      "content": "▁<extra_id_90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64270": {
+      "content": "▁<extra_id_89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64271": {
+      "content": "▁<extra_id_88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64272": {
+      "content": "▁<extra_id_87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64273": {
+      "content": "▁<extra_id_86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64274": {
+      "content": "▁<extra_id_85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64275": {
+      "content": "▁<extra_id_84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64276": {
+      "content": "▁<extra_id_83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64277": {
+      "content": "▁<extra_id_82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64278": {
+      "content": "▁<extra_id_81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64279": {
+      "content": "▁<extra_id_80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64280": {
+      "content": "▁<extra_id_79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64281": {
+      "content": "▁<extra_id_78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64282": {
+      "content": "▁<extra_id_77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64283": {
+      "content": "▁<extra_id_76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64284": {
+      "content": "▁<extra_id_75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64285": {
+      "content": "▁<extra_id_74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64286": {
+      "content": "▁<extra_id_73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64287": {
+      "content": "▁<extra_id_72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64288": {
+      "content": "▁<extra_id_71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64289": {
+      "content": "▁<extra_id_70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64290": {
+      "content": "▁<extra_id_69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64291": {
+      "content": "▁<extra_id_68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64292": {
+      "content": "▁<extra_id_67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64293": {
+      "content": "▁<extra_id_66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64294": {
+      "content": "▁<extra_id_65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64295": {
+      "content": "▁<extra_id_64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64296": {
+      "content": "▁<extra_id_63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64297": {
+      "content": "▁<extra_id_62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64298": {
+      "content": "▁<extra_id_61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64299": {
+      "content": "▁<extra_id_60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64300": {
+      "content": "▁<extra_id_59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64301": {
+      "content": "▁<extra_id_58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64302": {
+      "content": "▁<extra_id_57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64303": {
+      "content": "▁<extra_id_56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64304": {
+      "content": "▁<extra_id_55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64305": {
+      "content": "▁<extra_id_54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64306": {
+      "content": "▁<extra_id_53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64307": {
+      "content": "▁<extra_id_52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64308": {
+      "content": "▁<extra_id_51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64309": {
+      "content": "▁<extra_id_50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64310": {
+      "content": "▁<extra_id_49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64311": {
+      "content": "▁<extra_id_48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64312": {
+      "content": "▁<extra_id_47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64313": {
+      "content": "▁<extra_id_46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64314": {
+      "content": "▁<extra_id_45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64315": {
+      "content": "▁<extra_id_44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64316": {
+      "content": "▁<extra_id_43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64317": {
+      "content": "▁<extra_id_42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64318": {
+      "content": "▁<extra_id_41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64319": {
+      "content": "▁<extra_id_40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64320": {
+      "content": "▁<extra_id_39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64321": {
+      "content": "▁<extra_id_38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64322": {
+      "content": "▁<extra_id_37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64323": {
+      "content": "▁<extra_id_36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64324": {
+      "content": "▁<extra_id_35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64325": {
+      "content": "▁<extra_id_34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64326": {
+      "content": "▁<extra_id_33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64327": {
+      "content": "▁<extra_id_32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64328": {
+      "content": "▁<extra_id_31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64329": {
+      "content": "▁<extra_id_30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64330": {
+      "content": "▁<extra_id_29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64331": {
+      "content": "▁<extra_id_28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64332": {
+      "content": "▁<extra_id_27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64333": {
+      "content": "▁<extra_id_26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64334": {
+      "content": "▁<extra_id_25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64335": {
+      "content": "▁<extra_id_24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64336": {
+      "content": "▁<extra_id_23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64337": {
+      "content": "▁<extra_id_22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64338": {
+      "content": "▁<extra_id_21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64339": {
+      "content": "▁<extra_id_20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64340": {
+      "content": "▁<extra_id_19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64341": {
+      "content": "▁<extra_id_18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64342": {
+      "content": "▁<extra_id_17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64343": {
+      "content": "▁<extra_id_16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64344": {
+      "content": "▁<extra_id_15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64345": {
+      "content": "▁<extra_id_14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64346": {
+      "content": "▁<extra_id_13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64347": {
+      "content": "▁<extra_id_12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64348": {
+      "content": "▁<extra_id_11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64349": {
+      "content": "▁<extra_id_10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64350": {
+      "content": "▁<extra_id_9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64351": {
+      "content": "▁<extra_id_8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64352": {
+      "content": "▁<extra_id_7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64353": {
+      "content": "▁<extra_id_6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64354": {
+      "content": "▁<extra_id_5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64355": {
+      "content": "▁<extra_id_4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64356": {
+      "content": "▁<extra_id_3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64357": {
+      "content": "▁<extra_id_2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64358": {
+      "content": "▁<extra_id_1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64359": {
+      "content": "▁<extra_id_0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64360": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "[CLS]",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "[CLS]",
+  "do_lower_case": false,
+  "eos_token": "[SEP]",
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "sp_model_kwargs": {},
+  "split_by_punct": false,
+  "tokenizer_class": "DebertaV2Tokenizer",
+  "unk_token": "[UNK]"
+}