Hugging Face's logo

PuxAI
/
GenPII-framework-Test

Safetensors
Model card Files
xet
 Community
PuxAI commited on
Commit
ed1b371
·
verified ·
1 Parent(s): 3a96fcc

Upload folder using huggingface_hub

Browse files
Files changed (20) hide show
  1. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/added_tokens.json +3 -0
  2. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/added_tokens.json +3 -0
  3. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/config.json +81 -0
  4. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/model.safetensors +3 -0
  5. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/optimizer.pt +3 -0
  6. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/rng_state.pth +3 -0
  7. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/scheduler.pt +3 -0
  8. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/special_tokens_map.json +15 -0
  9. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/spm.model +3 -0
  10. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/tokenizer.json +0 -0
  11. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/tokenizer_config.json +60 -0
  12. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/trainer_state.json +118 -0
  13. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/training_args.bin +3 -0
  14. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/config.json +81 -0
  15. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/pytorch_model.bin +3 -0
  16. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/special_tokens_map.json +15 -0
  17. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/spm.model +3 -0
  18. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/tokenizer.json +0 -0
  19. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/tokenizer_config.json +60 -0
  20. open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/training_meta.json +31 -0
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "[MASK]": 128000
3
+ }
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "[MASK]": 128000
3
+ }
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/config.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DebertaAdvancedSpanClassifier"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "dtype": "float32",
7
+ "hidden_act": "gelu",
8
+ "hidden_dropout_prob": 0.1,
9
+ "hidden_size": 768,
10
+ "id2label": {
11
+ "0": "O",
12
+ "1": "AGE",
13
+ "2": "BUILDINGNUM",
14
+ "3": "CITY",
15
+ "4": "CREDITCARDNUMBER",
16
+ "5": "DATE",
17
+ "6": "DRIVERLICENSENUM",
18
+ "7": "EMAIL",
19
+ "8": "GENDER",
20
+ "9": "GIVENNAME",
21
+ "10": "IDCARDNUM",
22
+ "11": "PASSPORTNUM",
23
+ "12": "SEX",
24
+ "13": "SOCIALNUM",
25
+ "14": "STREET",
26
+ "15": "SURNAME",
27
+ "16": "TAXNUM",
28
+ "17": "TELEPHONENUM",
29
+ "18": "TIME",
30
+ "19": "TITLE",
31
+ "20": "ZIPCODE"
32
+ },
33
+ "initializer_range": 0.02,
34
+ "intermediate_size": 3072,
35
+ "label2id": {
36
+ "AGE": 1,
37
+ "BUILDINGNUM": 2,
38
+ "CITY": 3,
39
+ "CREDITCARDNUMBER": 4,
40
+ "DATE": 5,
41
+ "DRIVERLICENSENUM": 6,
42
+ "EMAIL": 7,
43
+ "GENDER": 8,
44
+ "GIVENNAME": 9,
45
+ "IDCARDNUM": 10,
46
+ "O": 0,
47
+ "PASSPORTNUM": 11,
48
+ "SEX": 12,
49
+ "SOCIALNUM": 13,
50
+ "STREET": 14,
51
+ "SURNAME": 15,
52
+ "TAXNUM": 16,
53
+ "TELEPHONENUM": 17,
54
+ "TIME": 18,
55
+ "TITLE": 19,
56
+ "ZIPCODE": 20
57
+ },
58
+ "layer_norm_eps": 1e-07,
59
+ "legacy": true,
60
+ "max_position_embeddings": 512,
61
+ "max_relative_positions": -1,
62
+ "model_type": "deberta-v2",
63
+ "norm_rel_ebd": "layer_norm",
64
+ "num_attention_heads": 12,
65
+ "num_hidden_layers": 12,
66
+ "pad_token_id": 0,
67
+ "pooler_dropout": 0,
68
+ "pooler_hidden_act": "gelu",
69
+ "pooler_hidden_size": 768,
70
+ "pos_att_type": [
71
+ "p2c",
72
+ "c2p"
73
+ ],
74
+ "position_biased_input": false,
75
+ "position_buckets": 256,
76
+ "relative_attention": true,
77
+ "share_att_key": true,
78
+ "transformers_version": "4.57.6",
79
+ "type_vocab_size": 0,
80
+ "vocab_size": 128100
81
+ }
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ada1c0f600f426bd305d70cea962d0250d8de1fee6f4fb70d13000d68a145f99
3
+ size 741362860
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5597b493068e0f4a52112cf2e923e89a1e35541304f4a9f654db281802f0ded8
3
+ size 1482849419
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce738eff120000031daad839b26f5ff0564cfc6338e9cd70aae83be58409c232
3
+ size 14645
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1f8ef2703772c350ae1ecab2e06f59a66e2bce06469bc2c7f1657f1366e4c7a
3
+ size 1465
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "[CLS]",
3
+ "cls_token": "[CLS]",
4
+ "eos_token": "[SEP]",
5
+ "mask_token": "[MASK]",
6
+ "pad_token": "[PAD]",
7
+ "sep_token": "[SEP]",
8
+ "unk_token": {
9
+ "content": "[UNK]",
10
+ "lstrip": false,
11
+ "normalized": true,
12
+ "rstrip": false,
13
+ "single_word": false
14
+ }
15
+ }
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/spm.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
3
+ size 2464616
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/tokenizer_config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "[PAD]",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "[CLS]",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "[SEP]",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "[UNK]",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "128000": {
37
+ "content": "[MASK]",
38
+ "lstrip": false,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "[CLS]",
46
+ "clean_up_tokenization_spaces": false,
47
+ "cls_token": "[CLS]",
48
+ "do_lower_case": false,
49
+ "eos_token": "[SEP]",
50
+ "extra_special_tokens": {},
51
+ "mask_token": "[MASK]",
52
+ "model_max_length": 1000000000000000019884624838656,
53
+ "pad_token": "[PAD]",
54
+ "sep_token": "[SEP]",
55
+ "sp_model_kwargs": {},
56
+ "split_by_punct": false,
57
+ "tokenizer_class": "DebertaV2Tokenizer",
58
+ "unk_token": "[UNK]",
59
+ "vocab_type": "spm"
60
+ }
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/trainer_state.json ADDED
@@ -0,0 +1,118 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 5.0,
6
+ "eval_steps": 500,
7
+ "global_step": 6395,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.39093041438623927,
14
+ "grad_norm": 0.2406579703092575,
15
+ "learning_rate": 3.11875e-05,
16
+ "loss": 0.51,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 0.7818608287724785,
21
+ "grad_norm": 0.1867370307445526,
22
+ "learning_rate": 3.7504778453518686e-05,
23
+ "loss": 0.03,
24
+ "step": 1000
25
+ },
26
+ {
27
+ "epoch": 1.1727912431587177,
28
+ "grad_norm": 0.11661892384290695,
29
+ "learning_rate": 3.402953953084275e-05,
30
+ "loss": 0.0215,
31
+ "step": 1500
32
+ },
33
+ {
34
+ "epoch": 1.5637216575449568,
35
+ "grad_norm": 0.11837314069271088,
36
+ "learning_rate": 3.055430060816682e-05,
37
+ "loss": 0.0181,
38
+ "step": 2000
39
+ },
40
+ {
41
+ "epoch": 1.9546520719311964,
42
+ "grad_norm": 0.11518122255802155,
43
+ "learning_rate": 2.7079061685490882e-05,
44
+ "loss": 0.0163,
45
+ "step": 2500
46
+ },
47
+ {
48
+ "epoch": 2.3455824863174355,
49
+ "grad_norm": 0.11215662211179733,
50
+ "learning_rate": 2.3603822762814948e-05,
51
+ "loss": 0.0139,
52
+ "step": 3000
53
+ },
54
+ {
55
+ "epoch": 2.7365129007036746,
56
+ "grad_norm": 0.15525175631046295,
57
+ "learning_rate": 2.0128583840139013e-05,
58
+ "loss": 0.0136,
59
+ "step": 3500
60
+ },
61
+ {
62
+ "epoch": 3.127443315089914,
63
+ "grad_norm": 0.11564604192972183,
64
+ "learning_rate": 1.665334491746308e-05,
65
+ "loss": 0.0125,
66
+ "step": 4000
67
+ },
68
+ {
69
+ "epoch": 3.5183737294761532,
70
+ "grad_norm": 0.10782811790704727,
71
+ "learning_rate": 1.3178105994787142e-05,
72
+ "loss": 0.0114,
73
+ "step": 4500
74
+ },
75
+ {
76
+ "epoch": 3.9093041438623923,
77
+ "grad_norm": 0.09704761207103729,
78
+ "learning_rate": 9.702867072111208e-06,
79
+ "loss": 0.0111,
80
+ "step": 5000
81
+ },
82
+ {
83
+ "epoch": 4.300234558248632,
84
+ "grad_norm": 0.1157800629734993,
85
+ "learning_rate": 6.227628149435274e-06,
86
+ "loss": 0.0102,
87
+ "step": 5500
88
+ },
89
+ {
90
+ "epoch": 4.691164972634871,
91
+ "grad_norm": 0.11194238066673279,
92
+ "learning_rate": 2.75238922675934e-06,
93
+ "loss": 0.0098,
94
+ "step": 6000
95
+ }
96
+ ],
97
+ "logging_steps": 500,
98
+ "max_steps": 6395,
99
+ "num_input_tokens_seen": 0,
100
+ "num_train_epochs": 5,
101
+ "save_steps": 500,
102
+ "stateful_callbacks": {
103
+ "TrainerControl": {
104
+ "args": {
105
+ "should_epoch_stop": false,
106
+ "should_evaluate": false,
107
+ "should_log": false,
108
+ "should_save": true,
109
+ "should_training_stop": true
110
+ },
111
+ "attributes": {}
112
+ }
113
+ },
114
+ "total_flos": 1.3339049651519789e+17,
115
+ "train_batch_size": 64,
116
+ "trial_name": null,
117
+ "trial_params": null
118
+ }
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/checkpoint-6395/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:833c743deb711201d1b0dbc5c81b611355b8fee48d478492eb0888b03f2cc93b
3
+ size 5905
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/config.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DebertaAdvancedSpanClassifier"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "dtype": "float32",
7
+ "hidden_act": "gelu",
8
+ "hidden_dropout_prob": 0.1,
9
+ "hidden_size": 768,
10
+ "id2label": {
11
+ "0": "O",
12
+ "1": "AGE",
13
+ "2": "BUILDINGNUM",
14
+ "3": "CITY",
15
+ "4": "CREDITCARDNUMBER",
16
+ "5": "DATE",
17
+ "6": "DRIVERLICENSENUM",
18
+ "7": "EMAIL",
19
+ "8": "GENDER",
20
+ "9": "GIVENNAME",
21
+ "10": "IDCARDNUM",
22
+ "11": "PASSPORTNUM",
23
+ "12": "SEX",
24
+ "13": "SOCIALNUM",
25
+ "14": "STREET",
26
+ "15": "SURNAME",
27
+ "16": "TAXNUM",
28
+ "17": "TELEPHONENUM",
29
+ "18": "TIME",
30
+ "19": "TITLE",
31
+ "20": "ZIPCODE"
32
+ },
33
+ "initializer_range": 0.02,
34
+ "intermediate_size": 3072,
35
+ "label2id": {
36
+ "AGE": 1,
37
+ "BUILDINGNUM": 2,
38
+ "CITY": 3,
39
+ "CREDITCARDNUMBER": 4,
40
+ "DATE": 5,
41
+ "DRIVERLICENSENUM": 6,
42
+ "EMAIL": 7,
43
+ "GENDER": 8,
44
+ "GIVENNAME": 9,
45
+ "IDCARDNUM": 10,
46
+ "O": 0,
47
+ "PASSPORTNUM": 11,
48
+ "SEX": 12,
49
+ "SOCIALNUM": 13,
50
+ "STREET": 14,
51
+ "SURNAME": 15,
52
+ "TAXNUM": 16,
53
+ "TELEPHONENUM": 17,
54
+ "TIME": 18,
55
+ "TITLE": 19,
56
+ "ZIPCODE": 20
57
+ },
58
+ "layer_norm_eps": 1e-07,
59
+ "legacy": true,
60
+ "max_position_embeddings": 512,
61
+ "max_relative_positions": -1,
62
+ "model_type": "deberta-v2",
63
+ "norm_rel_ebd": "layer_norm",
64
+ "num_attention_heads": 12,
65
+ "num_hidden_layers": 12,
66
+ "pad_token_id": 0,
67
+ "pooler_dropout": 0,
68
+ "pooler_hidden_act": "gelu",
69
+ "pooler_hidden_size": 768,
70
+ "pos_att_type": [
71
+ "p2c",
72
+ "c2p"
73
+ ],
74
+ "position_biased_input": false,
75
+ "position_buckets": 256,
76
+ "relative_attention": true,
77
+ "share_att_key": true,
78
+ "transformers_version": "4.57.6",
79
+ "type_vocab_size": 0,
80
+ "vocab_size": 128100
81
+ }
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1fdeb3f3333ca5df29eb703b28f36d0519ecb20aad51e3e22c1c1941a653b8d
3
+ size 741424783
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "[CLS]",
3
+ "cls_token": "[CLS]",
4
+ "eos_token": "[SEP]",
5
+ "mask_token": "[MASK]",
6
+ "pad_token": "[PAD]",
7
+ "sep_token": "[SEP]",
8
+ "unk_token": {
9
+ "content": "[UNK]",
10
+ "lstrip": false,
11
+ "normalized": true,
12
+ "rstrip": false,
13
+ "single_word": false
14
+ }
15
+ }
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/spm.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
3
+ size 2464616
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/tokenizer_config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "[PAD]",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "[CLS]",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "[SEP]",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "[UNK]",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "128000": {
37
+ "content": "[MASK]",
38
+ "lstrip": false,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "[CLS]",
46
+ "clean_up_tokenization_spaces": false,
47
+ "cls_token": "[CLS]",
48
+ "do_lower_case": false,
49
+ "eos_token": "[SEP]",
50
+ "extra_special_tokens": {},
51
+ "mask_token": "[MASK]",
52
+ "model_max_length": 1000000000000000019884624838656,
53
+ "pad_token": "[PAD]",
54
+ "sep_token": "[SEP]",
55
+ "sp_model_kwargs": {},
56
+ "split_by_punct": false,
57
+ "tokenizer_class": "DebertaV2Tokenizer",
58
+ "unk_token": "[UNK]",
59
+ "vocab_type": "spm"
60
+ }
open-pii-masking-500k-ai4privacy-augmented/span-deberta-v3-base/training_meta.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architecture": "span_based",
3
+ "base_model": "microsoft/deberta-v3-base",
4
+ "dataset": "open-pii-masking-500k-ai4privacy-augmented",
5
+ "label_list": [
6
+ "O",
7
+ "AGE",
8
+ "BUILDINGNUM",
9
+ "CITY",
10
+ "CREDITCARDNUMBER",
11
+ "DATE",
12
+ "DRIVERLICENSENUM",
13
+ "EMAIL",
14
+ "GENDER",
15
+ "GIVENNAME",
16
+ "IDCARDNUM",
17
+ "PASSPORTNUM",
18
+ "SEX",
19
+ "SOCIALNUM",
20
+ "STREET",
21
+ "SURNAME",
22
+ "TAXNUM",
23
+ "TELEPHONENUM",
24
+ "TIME",
25
+ "TITLE",
26
+ "ZIPCODE"
27
+ ],
28
+ "max_seq_length": 512,
29
+ "max_span_width": 30,
30
+ "timestamp": "2026-04-07T06:20:35.864254"
31
+ }