PuxAI commited on
Commit
15b5132
·
verified ·
1 Parent(s): 06aa18f

Upload folder using huggingface_hub

Browse files
nemotron-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/config.json ADDED
@@ -0,0 +1,153 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DebertaStandardSpanClassifier"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bos_token_id": null,
7
+ "dtype": "float32",
8
+ "eos_token_id": null,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "id2label": {
13
+ "0": "O",
14
+ "1": "ACCOUNT_NUMBER",
15
+ "2": "AGE",
16
+ "3": "API_KEY",
17
+ "4": "BANK_ROUTING_NUMBER",
18
+ "5": "BIOMETRIC_IDENTIFIER",
19
+ "6": "BLOOD_TYPE",
20
+ "7": "CERTIFICATE_LICENSE_NUMBER",
21
+ "8": "CITY",
22
+ "9": "COMPANY_NAME",
23
+ "10": "COORDINATE",
24
+ "11": "COUNTRY",
25
+ "12": "COUNTY",
26
+ "13": "CREDIT_DEBIT_CARD",
27
+ "14": "CUSTOMER_ID",
28
+ "15": "CVV",
29
+ "16": "DATE",
30
+ "17": "DATE_OF_BIRTH",
31
+ "18": "DATE_TIME",
32
+ "19": "DEVICE_IDENTIFIER",
33
+ "20": "EDUCATION_LEVEL",
34
+ "21": "EMAIL",
35
+ "22": "EMPLOYEE_ID",
36
+ "23": "EMPLOYMENT_STATUS",
37
+ "24": "FAX_NUMBER",
38
+ "25": "FIRST_NAME",
39
+ "26": "GENDER",
40
+ "27": "HEALTH_PLAN_BENEFICIARY_NUMBER",
41
+ "28": "HTTP_COOKIE",
42
+ "29": "IPV4",
43
+ "30": "IPV6",
44
+ "31": "LANGUAGE",
45
+ "32": "LAST_NAME",
46
+ "33": "LICENSE_PLATE",
47
+ "34": "MAC_ADDRESS",
48
+ "35": "MEDICAL_RECORD_NUMBER",
49
+ "36": "OCCUPATION",
50
+ "37": "PASSWORD",
51
+ "38": "PHONE_NUMBER",
52
+ "39": "PIN",
53
+ "40": "POLITICAL_VIEW",
54
+ "41": "POSTCODE",
55
+ "42": "RACE_ETHNICITY",
56
+ "43": "RELIGIOUS_BELIEF",
57
+ "44": "SEXUALITY",
58
+ "45": "SSN",
59
+ "46": "STATE",
60
+ "47": "STREET_ADDRESS",
61
+ "48": "SWIFT_BIC",
62
+ "49": "TAX_ID",
63
+ "50": "TIME",
64
+ "51": "UNIQUE_ID",
65
+ "52": "URL",
66
+ "53": "USER_NAME",
67
+ "54": "VEHICLE_IDENTIFIER"
68
+ },
69
+ "initializer_range": 0.02,
70
+ "intermediate_size": 3072,
71
+ "label2id": {
72
+ "ACCOUNT_NUMBER": 1,
73
+ "AGE": 2,
74
+ "API_KEY": 3,
75
+ "BANK_ROUTING_NUMBER": 4,
76
+ "BIOMETRIC_IDENTIFIER": 5,
77
+ "BLOOD_TYPE": 6,
78
+ "CERTIFICATE_LICENSE_NUMBER": 7,
79
+ "CITY": 8,
80
+ "COMPANY_NAME": 9,
81
+ "COORDINATE": 10,
82
+ "COUNTRY": 11,
83
+ "COUNTY": 12,
84
+ "CREDIT_DEBIT_CARD": 13,
85
+ "CUSTOMER_ID": 14,
86
+ "CVV": 15,
87
+ "DATE": 16,
88
+ "DATE_OF_BIRTH": 17,
89
+ "DATE_TIME": 18,
90
+ "DEVICE_IDENTIFIER": 19,
91
+ "EDUCATION_LEVEL": 20,
92
+ "EMAIL": 21,
93
+ "EMPLOYEE_ID": 22,
94
+ "EMPLOYMENT_STATUS": 23,
95
+ "FAX_NUMBER": 24,
96
+ "FIRST_NAME": 25,
97
+ "GENDER": 26,
98
+ "HEALTH_PLAN_BENEFICIARY_NUMBER": 27,
99
+ "HTTP_COOKIE": 28,
100
+ "IPV4": 29,
101
+ "IPV6": 30,
102
+ "LANGUAGE": 31,
103
+ "LAST_NAME": 32,
104
+ "LICENSE_PLATE": 33,
105
+ "MAC_ADDRESS": 34,
106
+ "MEDICAL_RECORD_NUMBER": 35,
107
+ "O": 0,
108
+ "OCCUPATION": 36,
109
+ "PASSWORD": 37,
110
+ "PHONE_NUMBER": 38,
111
+ "PIN": 39,
112
+ "POLITICAL_VIEW": 40,
113
+ "POSTCODE": 41,
114
+ "RACE_ETHNICITY": 42,
115
+ "RELIGIOUS_BELIEF": 43,
116
+ "SEXUALITY": 44,
117
+ "SSN": 45,
118
+ "STATE": 46,
119
+ "STREET_ADDRESS": 47,
120
+ "SWIFT_BIC": 48,
121
+ "TAX_ID": 49,
122
+ "TIME": 50,
123
+ "UNIQUE_ID": 51,
124
+ "URL": 52,
125
+ "USER_NAME": 53,
126
+ "VEHICLE_IDENTIFIER": 54
127
+ },
128
+ "layer_norm_eps": 1e-07,
129
+ "legacy": true,
130
+ "max_position_embeddings": 512,
131
+ "max_relative_positions": -1,
132
+ "model_type": "deberta-v2",
133
+ "norm_rel_ebd": "layer_norm",
134
+ "num_attention_heads": 12,
135
+ "num_hidden_layers": 12,
136
+ "pad_token_id": 0,
137
+ "pooler_dropout": 0,
138
+ "pooler_hidden_act": "gelu",
139
+ "pooler_hidden_size": 768,
140
+ "pos_att_type": [
141
+ "p2c",
142
+ "c2p"
143
+ ],
144
+ "position_biased_input": false,
145
+ "position_buckets": 256,
146
+ "relative_attention": true,
147
+ "share_att_key": true,
148
+ "tie_word_embeddings": true,
149
+ "transformers_version": "5.3.0",
150
+ "type_vocab_size": 0,
151
+ "use_cache": false,
152
+ "vocab_size": 128100
153
+ }
nemotron-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e3c4b7947d99e794e564a45bd7afc94dbd29a80b8519c840224139d27144ac0
3
+ size 735688732
nemotron-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f614c29386a9280f485b2cb0f2fe23b6484b01af8cc5a311655c7e8ba039683
3
+ size 1471499211
nemotron-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9abc555cf596a90a39f94d9dea1276a44b74ccf72a8d2d0169d74145d97a9111
3
+ size 14645
nemotron-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b04611bea80597dee677a377375d057cf3fa3437fb4ed8809d4662285b8811b
3
+ size 1465
nemotron-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
nemotron-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/tokenizer_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "[CLS]",
5
+ "cls_token": "[CLS]",
6
+ "do_lower_case": false,
7
+ "eos_token": "[SEP]",
8
+ "extra_special_tokens": [
9
+ "[PAD]",
10
+ "[CLS]",
11
+ "[SEP]"
12
+ ],
13
+ "is_local": false,
14
+ "mask_token": "[MASK]",
15
+ "model_max_length": 1000000000000000019884624838656,
16
+ "pad_token": "[PAD]",
17
+ "sep_token": "[SEP]",
18
+ "split_by_punct": false,
19
+ "tokenizer_class": "DebertaV2Tokenizer",
20
+ "unk_id": 3,
21
+ "unk_token": "[UNK]",
22
+ "vocab_type": "spm"
23
+ }
nemotron-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/trainer_state.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 250,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [],
12
+ "logging_steps": 500,
13
+ "max_steps": 250,
14
+ "num_input_tokens_seen": 0,
15
+ "num_train_epochs": 1,
16
+ "save_steps": 500,
17
+ "stateful_callbacks": {
18
+ "TrainerControl": {
19
+ "args": {
20
+ "should_epoch_stop": false,
21
+ "should_evaluate": false,
22
+ "should_log": false,
23
+ "should_save": true,
24
+ "should_training_stop": true
25
+ },
26
+ "attributes": {}
27
+ }
28
+ },
29
+ "total_flos": 67906591283328.0,
30
+ "train_batch_size": 2,
31
+ "trial_name": null,
32
+ "trial_params": null
33
+ }
nemotron-pii-ready/SpanBased-CustomDeBERTa/checkpoint-250/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:355808bbd100454b8151d1d70a81f30f986be5449db0181c06190951761b407e
3
+ size 5201
nemotron-pii-ready/SpanBased-CustomDeBERTa/config.json ADDED
@@ -0,0 +1,153 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DebertaStandardSpanClassifier"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bos_token_id": null,
7
+ "dtype": "float32",
8
+ "eos_token_id": null,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "id2label": {
13
+ "0": "O",
14
+ "1": "ACCOUNT_NUMBER",
15
+ "2": "AGE",
16
+ "3": "API_KEY",
17
+ "4": "BANK_ROUTING_NUMBER",
18
+ "5": "BIOMETRIC_IDENTIFIER",
19
+ "6": "BLOOD_TYPE",
20
+ "7": "CERTIFICATE_LICENSE_NUMBER",
21
+ "8": "CITY",
22
+ "9": "COMPANY_NAME",
23
+ "10": "COORDINATE",
24
+ "11": "COUNTRY",
25
+ "12": "COUNTY",
26
+ "13": "CREDIT_DEBIT_CARD",
27
+ "14": "CUSTOMER_ID",
28
+ "15": "CVV",
29
+ "16": "DATE",
30
+ "17": "DATE_OF_BIRTH",
31
+ "18": "DATE_TIME",
32
+ "19": "DEVICE_IDENTIFIER",
33
+ "20": "EDUCATION_LEVEL",
34
+ "21": "EMAIL",
35
+ "22": "EMPLOYEE_ID",
36
+ "23": "EMPLOYMENT_STATUS",
37
+ "24": "FAX_NUMBER",
38
+ "25": "FIRST_NAME",
39
+ "26": "GENDER",
40
+ "27": "HEALTH_PLAN_BENEFICIARY_NUMBER",
41
+ "28": "HTTP_COOKIE",
42
+ "29": "IPV4",
43
+ "30": "IPV6",
44
+ "31": "LANGUAGE",
45
+ "32": "LAST_NAME",
46
+ "33": "LICENSE_PLATE",
47
+ "34": "MAC_ADDRESS",
48
+ "35": "MEDICAL_RECORD_NUMBER",
49
+ "36": "OCCUPATION",
50
+ "37": "PASSWORD",
51
+ "38": "PHONE_NUMBER",
52
+ "39": "PIN",
53
+ "40": "POLITICAL_VIEW",
54
+ "41": "POSTCODE",
55
+ "42": "RACE_ETHNICITY",
56
+ "43": "RELIGIOUS_BELIEF",
57
+ "44": "SEXUALITY",
58
+ "45": "SSN",
59
+ "46": "STATE",
60
+ "47": "STREET_ADDRESS",
61
+ "48": "SWIFT_BIC",
62
+ "49": "TAX_ID",
63
+ "50": "TIME",
64
+ "51": "UNIQUE_ID",
65
+ "52": "URL",
66
+ "53": "USER_NAME",
67
+ "54": "VEHICLE_IDENTIFIER"
68
+ },
69
+ "initializer_range": 0.02,
70
+ "intermediate_size": 3072,
71
+ "label2id": {
72
+ "ACCOUNT_NUMBER": 1,
73
+ "AGE": 2,
74
+ "API_KEY": 3,
75
+ "BANK_ROUTING_NUMBER": 4,
76
+ "BIOMETRIC_IDENTIFIER": 5,
77
+ "BLOOD_TYPE": 6,
78
+ "CERTIFICATE_LICENSE_NUMBER": 7,
79
+ "CITY": 8,
80
+ "COMPANY_NAME": 9,
81
+ "COORDINATE": 10,
82
+ "COUNTRY": 11,
83
+ "COUNTY": 12,
84
+ "CREDIT_DEBIT_CARD": 13,
85
+ "CUSTOMER_ID": 14,
86
+ "CVV": 15,
87
+ "DATE": 16,
88
+ "DATE_OF_BIRTH": 17,
89
+ "DATE_TIME": 18,
90
+ "DEVICE_IDENTIFIER": 19,
91
+ "EDUCATION_LEVEL": 20,
92
+ "EMAIL": 21,
93
+ "EMPLOYEE_ID": 22,
94
+ "EMPLOYMENT_STATUS": 23,
95
+ "FAX_NUMBER": 24,
96
+ "FIRST_NAME": 25,
97
+ "GENDER": 26,
98
+ "HEALTH_PLAN_BENEFICIARY_NUMBER": 27,
99
+ "HTTP_COOKIE": 28,
100
+ "IPV4": 29,
101
+ "IPV6": 30,
102
+ "LANGUAGE": 31,
103
+ "LAST_NAME": 32,
104
+ "LICENSE_PLATE": 33,
105
+ "MAC_ADDRESS": 34,
106
+ "MEDICAL_RECORD_NUMBER": 35,
107
+ "O": 0,
108
+ "OCCUPATION": 36,
109
+ "PASSWORD": 37,
110
+ "PHONE_NUMBER": 38,
111
+ "PIN": 39,
112
+ "POLITICAL_VIEW": 40,
113
+ "POSTCODE": 41,
114
+ "RACE_ETHNICITY": 42,
115
+ "RELIGIOUS_BELIEF": 43,
116
+ "SEXUALITY": 44,
117
+ "SSN": 45,
118
+ "STATE": 46,
119
+ "STREET_ADDRESS": 47,
120
+ "SWIFT_BIC": 48,
121
+ "TAX_ID": 49,
122
+ "TIME": 50,
123
+ "UNIQUE_ID": 51,
124
+ "URL": 52,
125
+ "USER_NAME": 53,
126
+ "VEHICLE_IDENTIFIER": 54
127
+ },
128
+ "layer_norm_eps": 1e-07,
129
+ "legacy": true,
130
+ "max_position_embeddings": 512,
131
+ "max_relative_positions": -1,
132
+ "model_type": "deberta-v2",
133
+ "norm_rel_ebd": "layer_norm",
134
+ "num_attention_heads": 12,
135
+ "num_hidden_layers": 12,
136
+ "pad_token_id": 0,
137
+ "pooler_dropout": 0,
138
+ "pooler_hidden_act": "gelu",
139
+ "pooler_hidden_size": 768,
140
+ "pos_att_type": [
141
+ "p2c",
142
+ "c2p"
143
+ ],
144
+ "position_biased_input": false,
145
+ "position_buckets": 256,
146
+ "relative_attention": true,
147
+ "share_att_key": true,
148
+ "tie_word_embeddings": true,
149
+ "transformers_version": "5.3.0",
150
+ "type_vocab_size": 0,
151
+ "use_cache": false,
152
+ "vocab_size": 128100
153
+ }
nemotron-pii-ready/SpanBased-CustomDeBERTa/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:483e07a97f1af68383ecb17ce086e9bb80e8bbf2a7ae4813a509e185902a6391
3
+ size 735749827
nemotron-pii-ready/SpanBased-CustomDeBERTa/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
nemotron-pii-ready/SpanBased-CustomDeBERTa/tokenizer_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "[CLS]",
5
+ "cls_token": "[CLS]",
6
+ "do_lower_case": false,
7
+ "eos_token": "[SEP]",
8
+ "extra_special_tokens": [
9
+ "[PAD]",
10
+ "[CLS]",
11
+ "[SEP]"
12
+ ],
13
+ "is_local": false,
14
+ "mask_token": "[MASK]",
15
+ "model_max_length": 1000000000000000019884624838656,
16
+ "pad_token": "[PAD]",
17
+ "sep_token": "[SEP]",
18
+ "split_by_punct": false,
19
+ "tokenizer_class": "DebertaV2Tokenizer",
20
+ "unk_id": 3,
21
+ "unk_token": "[UNK]",
22
+ "vocab_type": "spm"
23
+ }