PuxAI commited on
Commit
ffc978a
·
verified ·
1 Parent(s): 69a8d18

Upload folder using huggingface_hub

Browse files
Internal-Run-EN/SpanBased-DeBERTa/checkpoint-125/config.json ADDED
@@ -0,0 +1,157 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DebertaAdvancedSpanClassifier"
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": "ACCOUNTNAME",
15
+ "2": "ACCOUNTNUMBER",
16
+ "3": "AGE",
17
+ "4": "AMOUNT",
18
+ "5": "BIC",
19
+ "6": "BITCOINADDRESS",
20
+ "7": "BUILDINGNUMBER",
21
+ "8": "CITY",
22
+ "9": "COMPANYNAME",
23
+ "10": "COUNTY",
24
+ "11": "CREDITCARDCVV",
25
+ "12": "CREDITCARDISSUER",
26
+ "13": "CREDITCARDNUMBER",
27
+ "14": "CURRENCY",
28
+ "15": "CURRENCYCODE",
29
+ "16": "CURRENCYNAME",
30
+ "17": "CURRENCYSYMBOL",
31
+ "18": "DATE",
32
+ "19": "DOB",
33
+ "20": "EMAIL",
34
+ "21": "ETHEREUMADDRESS",
35
+ "22": "EYECOLOR",
36
+ "23": "FIRSTNAME",
37
+ "24": "GENDER",
38
+ "25": "HEIGHT",
39
+ "26": "IBAN",
40
+ "27": "IP",
41
+ "28": "IPV4",
42
+ "29": "IPV6",
43
+ "30": "JOBAREA",
44
+ "31": "JOBTITLE",
45
+ "32": "JOBTYPE",
46
+ "33": "LASTNAME",
47
+ "34": "LITECOINADDRESS",
48
+ "35": "MAC",
49
+ "36": "MASKEDNUMBER",
50
+ "37": "MIDDLENAME",
51
+ "38": "NEARBYGPSCOORDINATE",
52
+ "39": "ORDINALDIRECTION",
53
+ "40": "PASSWORD",
54
+ "41": "PHONEIMEI",
55
+ "42": "PHONENUMBER",
56
+ "43": "PIN",
57
+ "44": "PREFIX",
58
+ "45": "SECONDARYADDRESS",
59
+ "46": "SEX",
60
+ "47": "SSN",
61
+ "48": "STATE",
62
+ "49": "STREET",
63
+ "50": "TIME",
64
+ "51": "URL",
65
+ "52": "USERAGENT",
66
+ "53": "USERNAME",
67
+ "54": "VEHICLEVIN",
68
+ "55": "VEHICLEVRM",
69
+ "56": "ZIPCODE"
70
+ },
71
+ "initializer_range": 0.02,
72
+ "intermediate_size": 3072,
73
+ "label2id": {
74
+ "ACCOUNTNAME": 1,
75
+ "ACCOUNTNUMBER": 2,
76
+ "AGE": 3,
77
+ "AMOUNT": 4,
78
+ "BIC": 5,
79
+ "BITCOINADDRESS": 6,
80
+ "BUILDINGNUMBER": 7,
81
+ "CITY": 8,
82
+ "COMPANYNAME": 9,
83
+ "COUNTY": 10,
84
+ "CREDITCARDCVV": 11,
85
+ "CREDITCARDISSUER": 12,
86
+ "CREDITCARDNUMBER": 13,
87
+ "CURRENCY": 14,
88
+ "CURRENCYCODE": 15,
89
+ "CURRENCYNAME": 16,
90
+ "CURRENCYSYMBOL": 17,
91
+ "DATE": 18,
92
+ "DOB": 19,
93
+ "EMAIL": 20,
94
+ "ETHEREUMADDRESS": 21,
95
+ "EYECOLOR": 22,
96
+ "FIRSTNAME": 23,
97
+ "GENDER": 24,
98
+ "HEIGHT": 25,
99
+ "IBAN": 26,
100
+ "IP": 27,
101
+ "IPV4": 28,
102
+ "IPV6": 29,
103
+ "JOBAREA": 30,
104
+ "JOBTITLE": 31,
105
+ "JOBTYPE": 32,
106
+ "LASTNAME": 33,
107
+ "LITECOINADDRESS": 34,
108
+ "MAC": 35,
109
+ "MASKEDNUMBER": 36,
110
+ "MIDDLENAME": 37,
111
+ "NEARBYGPSCOORDINATE": 38,
112
+ "O": 0,
113
+ "ORDINALDIRECTION": 39,
114
+ "PASSWORD": 40,
115
+ "PHONEIMEI": 41,
116
+ "PHONENUMBER": 42,
117
+ "PIN": 43,
118
+ "PREFIX": 44,
119
+ "SECONDARYADDRESS": 45,
120
+ "SEX": 46,
121
+ "SSN": 47,
122
+ "STATE": 48,
123
+ "STREET": 49,
124
+ "TIME": 50,
125
+ "URL": 51,
126
+ "USERAGENT": 52,
127
+ "USERNAME": 53,
128
+ "VEHICLEVIN": 54,
129
+ "VEHICLEVRM": 55,
130
+ "ZIPCODE": 56
131
+ },
132
+ "layer_norm_eps": 1e-07,
133
+ "legacy": true,
134
+ "max_position_embeddings": 512,
135
+ "max_relative_positions": -1,
136
+ "model_type": "deberta-v2",
137
+ "norm_rel_ebd": "layer_norm",
138
+ "num_attention_heads": 12,
139
+ "num_hidden_layers": 12,
140
+ "pad_token_id": 0,
141
+ "pooler_dropout": 0,
142
+ "pooler_hidden_act": "gelu",
143
+ "pooler_hidden_size": 768,
144
+ "pos_att_type": [
145
+ "p2c",
146
+ "c2p"
147
+ ],
148
+ "position_biased_input": false,
149
+ "position_buckets": 256,
150
+ "relative_attention": true,
151
+ "share_att_key": true,
152
+ "tie_word_embeddings": true,
153
+ "transformers_version": "5.3.0",
154
+ "type_vocab_size": 0,
155
+ "use_cache": false,
156
+ "vocab_size": 128100
157
+ }
Internal-Run-EN/SpanBased-DeBERTa/checkpoint-125/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bbee790176a9c91b51f4cc8fe9d7f93caec38f54562aafe58e806a9f30641283
3
+ size 741504316
Internal-Run-EN/SpanBased-DeBERTa/checkpoint-125/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b839189eb383b0153c6272122c1acda1f99495283f9adadae0403372d25d24f1
3
+ size 1483132299
Internal-Run-EN/SpanBased-DeBERTa/checkpoint-125/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af3be233fe863c3366e34b9bde0e8b13c470c0a2eccd2baced5048901759ba14
3
+ size 14645
Internal-Run-EN/SpanBased-DeBERTa/checkpoint-125/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb91409f6e91f963319fd0d44556f41f27633ebf23e20d40c7a5da1ccd610f15
3
+ size 1383
Internal-Run-EN/SpanBased-DeBERTa/checkpoint-125/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:770e85488b2d6f2b41a590e6c2bb599f3a174081e14b8c48d3638681b4a06597
3
+ size 1465
Internal-Run-EN/SpanBased-DeBERTa/checkpoint-125/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
Internal-Run-EN/SpanBased-DeBERTa/checkpoint-125/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
+ }
Internal-Run-EN/SpanBased-DeBERTa/checkpoint-125/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": 125,
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": 125,
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": 13181821578432.0,
30
+ "train_batch_size": 4,
31
+ "trial_name": null,
32
+ "trial_params": null
33
+ }
Internal-Run-EN/SpanBased-DeBERTa/checkpoint-125/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cd49d8d7796bf5429c49f9a054abc3dc92f0705a0720160453c9ec0b8f56d83
3
+ size 5201
Internal-Run-EN/SpanBased-DeBERTa/config.json ADDED
@@ -0,0 +1,157 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DebertaAdvancedSpanClassifier"
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": "ACCOUNTNAME",
15
+ "2": "ACCOUNTNUMBER",
16
+ "3": "AGE",
17
+ "4": "AMOUNT",
18
+ "5": "BIC",
19
+ "6": "BITCOINADDRESS",
20
+ "7": "BUILDINGNUMBER",
21
+ "8": "CITY",
22
+ "9": "COMPANYNAME",
23
+ "10": "COUNTY",
24
+ "11": "CREDITCARDCVV",
25
+ "12": "CREDITCARDISSUER",
26
+ "13": "CREDITCARDNUMBER",
27
+ "14": "CURRENCY",
28
+ "15": "CURRENCYCODE",
29
+ "16": "CURRENCYNAME",
30
+ "17": "CURRENCYSYMBOL",
31
+ "18": "DATE",
32
+ "19": "DOB",
33
+ "20": "EMAIL",
34
+ "21": "ETHEREUMADDRESS",
35
+ "22": "EYECOLOR",
36
+ "23": "FIRSTNAME",
37
+ "24": "GENDER",
38
+ "25": "HEIGHT",
39
+ "26": "IBAN",
40
+ "27": "IP",
41
+ "28": "IPV4",
42
+ "29": "IPV6",
43
+ "30": "JOBAREA",
44
+ "31": "JOBTITLE",
45
+ "32": "JOBTYPE",
46
+ "33": "LASTNAME",
47
+ "34": "LITECOINADDRESS",
48
+ "35": "MAC",
49
+ "36": "MASKEDNUMBER",
50
+ "37": "MIDDLENAME",
51
+ "38": "NEARBYGPSCOORDINATE",
52
+ "39": "ORDINALDIRECTION",
53
+ "40": "PASSWORD",
54
+ "41": "PHONEIMEI",
55
+ "42": "PHONENUMBER",
56
+ "43": "PIN",
57
+ "44": "PREFIX",
58
+ "45": "SECONDARYADDRESS",
59
+ "46": "SEX",
60
+ "47": "SSN",
61
+ "48": "STATE",
62
+ "49": "STREET",
63
+ "50": "TIME",
64
+ "51": "URL",
65
+ "52": "USERAGENT",
66
+ "53": "USERNAME",
67
+ "54": "VEHICLEVIN",
68
+ "55": "VEHICLEVRM",
69
+ "56": "ZIPCODE"
70
+ },
71
+ "initializer_range": 0.02,
72
+ "intermediate_size": 3072,
73
+ "label2id": {
74
+ "ACCOUNTNAME": 1,
75
+ "ACCOUNTNUMBER": 2,
76
+ "AGE": 3,
77
+ "AMOUNT": 4,
78
+ "BIC": 5,
79
+ "BITCOINADDRESS": 6,
80
+ "BUILDINGNUMBER": 7,
81
+ "CITY": 8,
82
+ "COMPANYNAME": 9,
83
+ "COUNTY": 10,
84
+ "CREDITCARDCVV": 11,
85
+ "CREDITCARDISSUER": 12,
86
+ "CREDITCARDNUMBER": 13,
87
+ "CURRENCY": 14,
88
+ "CURRENCYCODE": 15,
89
+ "CURRENCYNAME": 16,
90
+ "CURRENCYSYMBOL": 17,
91
+ "DATE": 18,
92
+ "DOB": 19,
93
+ "EMAIL": 20,
94
+ "ETHEREUMADDRESS": 21,
95
+ "EYECOLOR": 22,
96
+ "FIRSTNAME": 23,
97
+ "GENDER": 24,
98
+ "HEIGHT": 25,
99
+ "IBAN": 26,
100
+ "IP": 27,
101
+ "IPV4": 28,
102
+ "IPV6": 29,
103
+ "JOBAREA": 30,
104
+ "JOBTITLE": 31,
105
+ "JOBTYPE": 32,
106
+ "LASTNAME": 33,
107
+ "LITECOINADDRESS": 34,
108
+ "MAC": 35,
109
+ "MASKEDNUMBER": 36,
110
+ "MIDDLENAME": 37,
111
+ "NEARBYGPSCOORDINATE": 38,
112
+ "O": 0,
113
+ "ORDINALDIRECTION": 39,
114
+ "PASSWORD": 40,
115
+ "PHONEIMEI": 41,
116
+ "PHONENUMBER": 42,
117
+ "PIN": 43,
118
+ "PREFIX": 44,
119
+ "SECONDARYADDRESS": 45,
120
+ "SEX": 46,
121
+ "SSN": 47,
122
+ "STATE": 48,
123
+ "STREET": 49,
124
+ "TIME": 50,
125
+ "URL": 51,
126
+ "USERAGENT": 52,
127
+ "USERNAME": 53,
128
+ "VEHICLEVIN": 54,
129
+ "VEHICLEVRM": 55,
130
+ "ZIPCODE": 56
131
+ },
132
+ "layer_norm_eps": 1e-07,
133
+ "legacy": true,
134
+ "max_position_embeddings": 512,
135
+ "max_relative_positions": -1,
136
+ "model_type": "deberta-v2",
137
+ "norm_rel_ebd": "layer_norm",
138
+ "num_attention_heads": 12,
139
+ "num_hidden_layers": 12,
140
+ "pad_token_id": 0,
141
+ "pooler_dropout": 0,
142
+ "pooler_hidden_act": "gelu",
143
+ "pooler_hidden_size": 768,
144
+ "pos_att_type": [
145
+ "p2c",
146
+ "c2p"
147
+ ],
148
+ "position_biased_input": false,
149
+ "position_buckets": 256,
150
+ "relative_attention": true,
151
+ "share_att_key": true,
152
+ "tie_word_embeddings": true,
153
+ "transformers_version": "5.3.0",
154
+ "type_vocab_size": 0,
155
+ "use_cache": false,
156
+ "vocab_size": 128100
157
+ }
Internal-Run-EN/SpanBased-DeBERTa/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c139ce6e098c2a976524ea68bc2e666b12d1c00c3d4d157ba7bc0e43574e829
3
+ size 741566223
Internal-Run-EN/SpanBased-DeBERTa/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
Internal-Run-EN/SpanBased-DeBERTa/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
+ }