Upload folder using huggingface_hub
Browse files- nemotron-pii-ready-augmented/span-deberta-v3-base/added_tokens.json +3 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/added_tokens.json +3 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/config.json +151 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/model.safetensors +3 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/optimizer.pt +3 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/rng_state.pth +3 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/scheduler.pt +3 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/special_tokens_map.json +15 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/spm.model +3 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/tokenizer.json +0 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/tokenizer_config.json +60 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/trainer_state.json +62 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/training_args.bin +3 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/config.json +151 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/pytorch_model.bin +3 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/special_tokens_map.json +15 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/spm.model +3 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/tokenizer.json +0 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/tokenizer_config.json +60 -0
- nemotron-pii-ready-augmented/span-deberta-v3-base/training_meta.json +66 -0
nemotron-pii-ready-augmented/span-deberta-v3-base/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[MASK]": 128000
|
| 3 |
+
}
|
nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[MASK]": 128000
|
| 3 |
+
}
|
nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/config.json
ADDED
|
@@ -0,0 +1,151 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "ACCOUNT_NUMBER",
|
| 13 |
+
"2": "AGE",
|
| 14 |
+
"3": "API_KEY",
|
| 15 |
+
"4": "BANK_ROUTING_NUMBER",
|
| 16 |
+
"5": "BIOMETRIC_IDENTIFIER",
|
| 17 |
+
"6": "BLOOD_TYPE",
|
| 18 |
+
"7": "CERTIFICATE_LICENSE_NUMBER",
|
| 19 |
+
"8": "CITY",
|
| 20 |
+
"9": "COMPANY_NAME",
|
| 21 |
+
"10": "COORDINATE",
|
| 22 |
+
"11": "COUNTRY",
|
| 23 |
+
"12": "COUNTY",
|
| 24 |
+
"13": "CREDIT_DEBIT_CARD",
|
| 25 |
+
"14": "CUSTOMER_ID",
|
| 26 |
+
"15": "CVV",
|
| 27 |
+
"16": "DATE",
|
| 28 |
+
"17": "DATE_OF_BIRTH",
|
| 29 |
+
"18": "DATE_TIME",
|
| 30 |
+
"19": "DEVICE_IDENTIFIER",
|
| 31 |
+
"20": "EDUCATION_LEVEL",
|
| 32 |
+
"21": "EMAIL",
|
| 33 |
+
"22": "EMPLOYEE_ID",
|
| 34 |
+
"23": "EMPLOYMENT_STATUS",
|
| 35 |
+
"24": "FAX_NUMBER",
|
| 36 |
+
"25": "FIRST_NAME",
|
| 37 |
+
"26": "GENDER",
|
| 38 |
+
"27": "HEALTH_PLAN_BENEFICIARY_NUMBER",
|
| 39 |
+
"28": "HTTP_COOKIE",
|
| 40 |
+
"29": "IPV4",
|
| 41 |
+
"30": "IPV6",
|
| 42 |
+
"31": "LANGUAGE",
|
| 43 |
+
"32": "LAST_NAME",
|
| 44 |
+
"33": "LICENSE_PLATE",
|
| 45 |
+
"34": "MAC_ADDRESS",
|
| 46 |
+
"35": "MEDICAL_RECORD_NUMBER",
|
| 47 |
+
"36": "NATIONAL_ID",
|
| 48 |
+
"37": "OCCUPATION",
|
| 49 |
+
"38": "PASSWORD",
|
| 50 |
+
"39": "PHONE_NUMBER",
|
| 51 |
+
"40": "PIN",
|
| 52 |
+
"41": "POLITICAL_VIEW",
|
| 53 |
+
"42": "POSTCODE",
|
| 54 |
+
"43": "RACE_ETHNICITY",
|
| 55 |
+
"44": "RELIGIOUS_BELIEF",
|
| 56 |
+
"45": "SEXUALITY",
|
| 57 |
+
"46": "SSN",
|
| 58 |
+
"47": "STATE",
|
| 59 |
+
"48": "STREET_ADDRESS",
|
| 60 |
+
"49": "SWIFT_BIC",
|
| 61 |
+
"50": "TAX_ID",
|
| 62 |
+
"51": "TIME",
|
| 63 |
+
"52": "UNIQUE_ID",
|
| 64 |
+
"53": "URL",
|
| 65 |
+
"54": "USER_NAME",
|
| 66 |
+
"55": "VEHICLE_IDENTIFIER"
|
| 67 |
+
},
|
| 68 |
+
"initializer_range": 0.02,
|
| 69 |
+
"intermediate_size": 3072,
|
| 70 |
+
"label2id": {
|
| 71 |
+
"ACCOUNT_NUMBER": 1,
|
| 72 |
+
"AGE": 2,
|
| 73 |
+
"API_KEY": 3,
|
| 74 |
+
"BANK_ROUTING_NUMBER": 4,
|
| 75 |
+
"BIOMETRIC_IDENTIFIER": 5,
|
| 76 |
+
"BLOOD_TYPE": 6,
|
| 77 |
+
"CERTIFICATE_LICENSE_NUMBER": 7,
|
| 78 |
+
"CITY": 8,
|
| 79 |
+
"COMPANY_NAME": 9,
|
| 80 |
+
"COORDINATE": 10,
|
| 81 |
+
"COUNTRY": 11,
|
| 82 |
+
"COUNTY": 12,
|
| 83 |
+
"CREDIT_DEBIT_CARD": 13,
|
| 84 |
+
"CUSTOMER_ID": 14,
|
| 85 |
+
"CVV": 15,
|
| 86 |
+
"DATE": 16,
|
| 87 |
+
"DATE_OF_BIRTH": 17,
|
| 88 |
+
"DATE_TIME": 18,
|
| 89 |
+
"DEVICE_IDENTIFIER": 19,
|
| 90 |
+
"EDUCATION_LEVEL": 20,
|
| 91 |
+
"EMAIL": 21,
|
| 92 |
+
"EMPLOYEE_ID": 22,
|
| 93 |
+
"EMPLOYMENT_STATUS": 23,
|
| 94 |
+
"FAX_NUMBER": 24,
|
| 95 |
+
"FIRST_NAME": 25,
|
| 96 |
+
"GENDER": 26,
|
| 97 |
+
"HEALTH_PLAN_BENEFICIARY_NUMBER": 27,
|
| 98 |
+
"HTTP_COOKIE": 28,
|
| 99 |
+
"IPV4": 29,
|
| 100 |
+
"IPV6": 30,
|
| 101 |
+
"LANGUAGE": 31,
|
| 102 |
+
"LAST_NAME": 32,
|
| 103 |
+
"LICENSE_PLATE": 33,
|
| 104 |
+
"MAC_ADDRESS": 34,
|
| 105 |
+
"MEDICAL_RECORD_NUMBER": 35,
|
| 106 |
+
"NATIONAL_ID": 36,
|
| 107 |
+
"O": 0,
|
| 108 |
+
"OCCUPATION": 37,
|
| 109 |
+
"PASSWORD": 38,
|
| 110 |
+
"PHONE_NUMBER": 39,
|
| 111 |
+
"PIN": 40,
|
| 112 |
+
"POLITICAL_VIEW": 41,
|
| 113 |
+
"POSTCODE": 42,
|
| 114 |
+
"RACE_ETHNICITY": 43,
|
| 115 |
+
"RELIGIOUS_BELIEF": 44,
|
| 116 |
+
"SEXUALITY": 45,
|
| 117 |
+
"SSN": 46,
|
| 118 |
+
"STATE": 47,
|
| 119 |
+
"STREET_ADDRESS": 48,
|
| 120 |
+
"SWIFT_BIC": 49,
|
| 121 |
+
"TAX_ID": 50,
|
| 122 |
+
"TIME": 51,
|
| 123 |
+
"UNIQUE_ID": 52,
|
| 124 |
+
"URL": 53,
|
| 125 |
+
"USER_NAME": 54,
|
| 126 |
+
"VEHICLE_IDENTIFIER": 55
|
| 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 |
+
"transformers_version": "4.57.6",
|
| 149 |
+
"type_vocab_size": 0,
|
| 150 |
+
"vocab_size": 128100
|
| 151 |
+
}
|
nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4fd130d63d271924519039cf9bedd2e9164de36b5d3e243d83913184ea821c38
|
| 3 |
+
size 741470520
|
nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c8ef79fe4ceeba5648b25e7400ad8c8023617caa213630b1483fe0f3374a8b42
|
| 3 |
+
size 1483064715
|
nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c829f8fbb0acb4d55ed0a74d9c257288c575054ffcfeec1a5656426fd672b4ae
|
| 3 |
+
size 14645
|
nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2df600276d717a293dd1e238e1c5b374910cdbbbc5dfbecfc3335f26071665cd
|
| 3 |
+
size 1465
|
nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/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 |
+
}
|
nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/spm.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
|
| 3 |
+
size 2464616
|
nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/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 |
+
}
|
nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/trainer_state.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 2285,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 1.0941262313024445,
|
| 14 |
+
"grad_norm": 0.06564472615718842,
|
| 15 |
+
"learning_rate": 3.474708171206226e-05,
|
| 16 |
+
"loss": 0.4622,
|
| 17 |
+
"step": 500
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 2.188252462604889,
|
| 21 |
+
"grad_norm": 0.07365531474351883,
|
| 22 |
+
"learning_rate": 2.501945525291829e-05,
|
| 23 |
+
"loss": 0.0132,
|
| 24 |
+
"step": 1000
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 3.282378693907333,
|
| 28 |
+
"grad_norm": 0.05786935240030289,
|
| 29 |
+
"learning_rate": 1.529182879377432e-05,
|
| 30 |
+
"loss": 0.0091,
|
| 31 |
+
"step": 1500
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 4.376504925209778,
|
| 35 |
+
"grad_norm": 0.04307584837079048,
|
| 36 |
+
"learning_rate": 5.564202334630351e-06,
|
| 37 |
+
"loss": 0.0071,
|
| 38 |
+
"step": 2000
|
| 39 |
+
}
|
| 40 |
+
],
|
| 41 |
+
"logging_steps": 500,
|
| 42 |
+
"max_steps": 2285,
|
| 43 |
+
"num_input_tokens_seen": 0,
|
| 44 |
+
"num_train_epochs": 5,
|
| 45 |
+
"save_steps": 500,
|
| 46 |
+
"stateful_callbacks": {
|
| 47 |
+
"TrainerControl": {
|
| 48 |
+
"args": {
|
| 49 |
+
"should_epoch_stop": false,
|
| 50 |
+
"should_evaluate": false,
|
| 51 |
+
"should_log": false,
|
| 52 |
+
"should_save": true,
|
| 53 |
+
"should_training_stop": true
|
| 54 |
+
},
|
| 55 |
+
"attributes": {}
|
| 56 |
+
}
|
| 57 |
+
},
|
| 58 |
+
"total_flos": 2.3321121457755034e+17,
|
| 59 |
+
"train_batch_size": 64,
|
| 60 |
+
"trial_name": null,
|
| 61 |
+
"trial_params": null
|
| 62 |
+
}
|
nemotron-pii-ready-augmented/span-deberta-v3-base/checkpoint-2285/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:35df3d888a1606dc5865415fa501b7bc756e0e8730b924dbf144dec2edb627c1
|
| 3 |
+
size 5905
|
nemotron-pii-ready-augmented/span-deberta-v3-base/config.json
ADDED
|
@@ -0,0 +1,151 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "ACCOUNT_NUMBER",
|
| 13 |
+
"2": "AGE",
|
| 14 |
+
"3": "API_KEY",
|
| 15 |
+
"4": "BANK_ROUTING_NUMBER",
|
| 16 |
+
"5": "BIOMETRIC_IDENTIFIER",
|
| 17 |
+
"6": "BLOOD_TYPE",
|
| 18 |
+
"7": "CERTIFICATE_LICENSE_NUMBER",
|
| 19 |
+
"8": "CITY",
|
| 20 |
+
"9": "COMPANY_NAME",
|
| 21 |
+
"10": "COORDINATE",
|
| 22 |
+
"11": "COUNTRY",
|
| 23 |
+
"12": "COUNTY",
|
| 24 |
+
"13": "CREDIT_DEBIT_CARD",
|
| 25 |
+
"14": "CUSTOMER_ID",
|
| 26 |
+
"15": "CVV",
|
| 27 |
+
"16": "DATE",
|
| 28 |
+
"17": "DATE_OF_BIRTH",
|
| 29 |
+
"18": "DATE_TIME",
|
| 30 |
+
"19": "DEVICE_IDENTIFIER",
|
| 31 |
+
"20": "EDUCATION_LEVEL",
|
| 32 |
+
"21": "EMAIL",
|
| 33 |
+
"22": "EMPLOYEE_ID",
|
| 34 |
+
"23": "EMPLOYMENT_STATUS",
|
| 35 |
+
"24": "FAX_NUMBER",
|
| 36 |
+
"25": "FIRST_NAME",
|
| 37 |
+
"26": "GENDER",
|
| 38 |
+
"27": "HEALTH_PLAN_BENEFICIARY_NUMBER",
|
| 39 |
+
"28": "HTTP_COOKIE",
|
| 40 |
+
"29": "IPV4",
|
| 41 |
+
"30": "IPV6",
|
| 42 |
+
"31": "LANGUAGE",
|
| 43 |
+
"32": "LAST_NAME",
|
| 44 |
+
"33": "LICENSE_PLATE",
|
| 45 |
+
"34": "MAC_ADDRESS",
|
| 46 |
+
"35": "MEDICAL_RECORD_NUMBER",
|
| 47 |
+
"36": "NATIONAL_ID",
|
| 48 |
+
"37": "OCCUPATION",
|
| 49 |
+
"38": "PASSWORD",
|
| 50 |
+
"39": "PHONE_NUMBER",
|
| 51 |
+
"40": "PIN",
|
| 52 |
+
"41": "POLITICAL_VIEW",
|
| 53 |
+
"42": "POSTCODE",
|
| 54 |
+
"43": "RACE_ETHNICITY",
|
| 55 |
+
"44": "RELIGIOUS_BELIEF",
|
| 56 |
+
"45": "SEXUALITY",
|
| 57 |
+
"46": "SSN",
|
| 58 |
+
"47": "STATE",
|
| 59 |
+
"48": "STREET_ADDRESS",
|
| 60 |
+
"49": "SWIFT_BIC",
|
| 61 |
+
"50": "TAX_ID",
|
| 62 |
+
"51": "TIME",
|
| 63 |
+
"52": "UNIQUE_ID",
|
| 64 |
+
"53": "URL",
|
| 65 |
+
"54": "USER_NAME",
|
| 66 |
+
"55": "VEHICLE_IDENTIFIER"
|
| 67 |
+
},
|
| 68 |
+
"initializer_range": 0.02,
|
| 69 |
+
"intermediate_size": 3072,
|
| 70 |
+
"label2id": {
|
| 71 |
+
"ACCOUNT_NUMBER": 1,
|
| 72 |
+
"AGE": 2,
|
| 73 |
+
"API_KEY": 3,
|
| 74 |
+
"BANK_ROUTING_NUMBER": 4,
|
| 75 |
+
"BIOMETRIC_IDENTIFIER": 5,
|
| 76 |
+
"BLOOD_TYPE": 6,
|
| 77 |
+
"CERTIFICATE_LICENSE_NUMBER": 7,
|
| 78 |
+
"CITY": 8,
|
| 79 |
+
"COMPANY_NAME": 9,
|
| 80 |
+
"COORDINATE": 10,
|
| 81 |
+
"COUNTRY": 11,
|
| 82 |
+
"COUNTY": 12,
|
| 83 |
+
"CREDIT_DEBIT_CARD": 13,
|
| 84 |
+
"CUSTOMER_ID": 14,
|
| 85 |
+
"CVV": 15,
|
| 86 |
+
"DATE": 16,
|
| 87 |
+
"DATE_OF_BIRTH": 17,
|
| 88 |
+
"DATE_TIME": 18,
|
| 89 |
+
"DEVICE_IDENTIFIER": 19,
|
| 90 |
+
"EDUCATION_LEVEL": 20,
|
| 91 |
+
"EMAIL": 21,
|
| 92 |
+
"EMPLOYEE_ID": 22,
|
| 93 |
+
"EMPLOYMENT_STATUS": 23,
|
| 94 |
+
"FAX_NUMBER": 24,
|
| 95 |
+
"FIRST_NAME": 25,
|
| 96 |
+
"GENDER": 26,
|
| 97 |
+
"HEALTH_PLAN_BENEFICIARY_NUMBER": 27,
|
| 98 |
+
"HTTP_COOKIE": 28,
|
| 99 |
+
"IPV4": 29,
|
| 100 |
+
"IPV6": 30,
|
| 101 |
+
"LANGUAGE": 31,
|
| 102 |
+
"LAST_NAME": 32,
|
| 103 |
+
"LICENSE_PLATE": 33,
|
| 104 |
+
"MAC_ADDRESS": 34,
|
| 105 |
+
"MEDICAL_RECORD_NUMBER": 35,
|
| 106 |
+
"NATIONAL_ID": 36,
|
| 107 |
+
"O": 0,
|
| 108 |
+
"OCCUPATION": 37,
|
| 109 |
+
"PASSWORD": 38,
|
| 110 |
+
"PHONE_NUMBER": 39,
|
| 111 |
+
"PIN": 40,
|
| 112 |
+
"POLITICAL_VIEW": 41,
|
| 113 |
+
"POSTCODE": 42,
|
| 114 |
+
"RACE_ETHNICITY": 43,
|
| 115 |
+
"RELIGIOUS_BELIEF": 44,
|
| 116 |
+
"SEXUALITY": 45,
|
| 117 |
+
"SSN": 46,
|
| 118 |
+
"STATE": 47,
|
| 119 |
+
"STREET_ADDRESS": 48,
|
| 120 |
+
"SWIFT_BIC": 49,
|
| 121 |
+
"TAX_ID": 50,
|
| 122 |
+
"TIME": 51,
|
| 123 |
+
"UNIQUE_ID": 52,
|
| 124 |
+
"URL": 53,
|
| 125 |
+
"USER_NAME": 54,
|
| 126 |
+
"VEHICLE_IDENTIFIER": 55
|
| 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 |
+
"transformers_version": "4.57.6",
|
| 149 |
+
"type_vocab_size": 0,
|
| 150 |
+
"vocab_size": 128100
|
| 151 |
+
}
|
nemotron-pii-ready-augmented/span-deberta-v3-base/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d568e93aefc685ab3577752df1174d9dd76a2a40c0fa753f3539e2877f9eaf3
|
| 3 |
+
size 741532431
|
nemotron-pii-ready-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 |
+
}
|
nemotron-pii-ready-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
|
nemotron-pii-ready-augmented/span-deberta-v3-base/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
nemotron-pii-ready-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 |
+
}
|
nemotron-pii-ready-augmented/span-deberta-v3-base/training_meta.json
ADDED
|
@@ -0,0 +1,66 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architecture": "span_based",
|
| 3 |
+
"base_model": "microsoft/deberta-v3-base",
|
| 4 |
+
"dataset": "nemotron-pii-ready-augmented",
|
| 5 |
+
"label_list": [
|
| 6 |
+
"O",
|
| 7 |
+
"ACCOUNT_NUMBER",
|
| 8 |
+
"AGE",
|
| 9 |
+
"API_KEY",
|
| 10 |
+
"BANK_ROUTING_NUMBER",
|
| 11 |
+
"BIOMETRIC_IDENTIFIER",
|
| 12 |
+
"BLOOD_TYPE",
|
| 13 |
+
"CERTIFICATE_LICENSE_NUMBER",
|
| 14 |
+
"CITY",
|
| 15 |
+
"COMPANY_NAME",
|
| 16 |
+
"COORDINATE",
|
| 17 |
+
"COUNTRY",
|
| 18 |
+
"COUNTY",
|
| 19 |
+
"CREDIT_DEBIT_CARD",
|
| 20 |
+
"CUSTOMER_ID",
|
| 21 |
+
"CVV",
|
| 22 |
+
"DATE",
|
| 23 |
+
"DATE_OF_BIRTH",
|
| 24 |
+
"DATE_TIME",
|
| 25 |
+
"DEVICE_IDENTIFIER",
|
| 26 |
+
"EDUCATION_LEVEL",
|
| 27 |
+
"EMAIL",
|
| 28 |
+
"EMPLOYEE_ID",
|
| 29 |
+
"EMPLOYMENT_STATUS",
|
| 30 |
+
"FAX_NUMBER",
|
| 31 |
+
"FIRST_NAME",
|
| 32 |
+
"GENDER",
|
| 33 |
+
"HEALTH_PLAN_BENEFICIARY_NUMBER",
|
| 34 |
+
"HTTP_COOKIE",
|
| 35 |
+
"IPV4",
|
| 36 |
+
"IPV6",
|
| 37 |
+
"LANGUAGE",
|
| 38 |
+
"LAST_NAME",
|
| 39 |
+
"LICENSE_PLATE",
|
| 40 |
+
"MAC_ADDRESS",
|
| 41 |
+
"MEDICAL_RECORD_NUMBER",
|
| 42 |
+
"NATIONAL_ID",
|
| 43 |
+
"OCCUPATION",
|
| 44 |
+
"PASSWORD",
|
| 45 |
+
"PHONE_NUMBER",
|
| 46 |
+
"PIN",
|
| 47 |
+
"POLITICAL_VIEW",
|
| 48 |
+
"POSTCODE",
|
| 49 |
+
"RACE_ETHNICITY",
|
| 50 |
+
"RELIGIOUS_BELIEF",
|
| 51 |
+
"SEXUALITY",
|
| 52 |
+
"SSN",
|
| 53 |
+
"STATE",
|
| 54 |
+
"STREET_ADDRESS",
|
| 55 |
+
"SWIFT_BIC",
|
| 56 |
+
"TAX_ID",
|
| 57 |
+
"TIME",
|
| 58 |
+
"UNIQUE_ID",
|
| 59 |
+
"URL",
|
| 60 |
+
"USER_NAME",
|
| 61 |
+
"VEHICLE_IDENTIFIER"
|
| 62 |
+
],
|
| 63 |
+
"max_seq_length": 512,
|
| 64 |
+
"max_span_width": 30,
|
| 65 |
+
"timestamp": "2026-04-07T09:21:07.401162"
|
| 66 |
+
}
|