Upload folder using huggingface_hub
Browse files- gretel-pii-ready-augmented/span-deberta-v3-base/added_tokens.json +3 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/added_tokens.json +3 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/config.json +125 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/model.safetensors +3 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/optimizer.pt +3 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/rng_state.pth +3 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/scheduler.pt +3 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/special_tokens_map.json +15 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/spm.model +3 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/tokenizer.json +0 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/tokenizer_config.json +60 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/trainer_state.json +41 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/training_args.bin +3 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/config.json +125 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/pytorch_model.bin +3 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/special_tokens_map.json +15 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/spm.model +3 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/tokenizer.json +0 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/tokenizer_config.json +60 -0
- gretel-pii-ready-augmented/span-deberta-v3-base/training_meta.json +53 -0
gretel-pii-ready-augmented/span-deberta-v3-base/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[MASK]": 128000
|
| 3 |
+
}
|
gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[MASK]": 128000
|
| 3 |
+
}
|
gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/config.json
ADDED
|
@@ -0,0 +1,125 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "ADDRESS",
|
| 14 |
+
"3": "API_KEY",
|
| 15 |
+
"4": "BANK_ROUTING_NUMBER",
|
| 16 |
+
"5": "BIOMETRIC_IDENTIFIER",
|
| 17 |
+
"6": "CERTIFICATE_LICENSE_NUMBER",
|
| 18 |
+
"7": "CITY",
|
| 19 |
+
"8": "COMPANY_NAME",
|
| 20 |
+
"9": "COORDINATE",
|
| 21 |
+
"10": "COUNTRY",
|
| 22 |
+
"11": "CREDIT_CARD_NUMBER",
|
| 23 |
+
"12": "CUSTOMER_ID",
|
| 24 |
+
"13": "CVV",
|
| 25 |
+
"14": "DATE",
|
| 26 |
+
"15": "DATE_OF_BIRTH",
|
| 27 |
+
"16": "DATE_TIME",
|
| 28 |
+
"17": "DEVICE_IDENTIFIER",
|
| 29 |
+
"18": "EMAIL",
|
| 30 |
+
"19": "EMPLOYEE_ID",
|
| 31 |
+
"20": "FIRST_NAME",
|
| 32 |
+
"21": "HEALTH_PLAN_BENEFICIARY_NUMBER",
|
| 33 |
+
"22": "IPV4",
|
| 34 |
+
"23": "IPV6",
|
| 35 |
+
"24": "LAST_NAME",
|
| 36 |
+
"25": "LICENSE_PLATE",
|
| 37 |
+
"26": "MEDICAL_RECORD_NUMBER",
|
| 38 |
+
"27": "NAME",
|
| 39 |
+
"28": "NATIONAL_ID",
|
| 40 |
+
"29": "PASSWORD",
|
| 41 |
+
"30": "PHONE_NUMBER",
|
| 42 |
+
"31": "PIN",
|
| 43 |
+
"32": "POSTCODE",
|
| 44 |
+
"33": "SSN",
|
| 45 |
+
"34": "STATE",
|
| 46 |
+
"35": "STREET_ADDRESS",
|
| 47 |
+
"36": "SWIFT_BIC",
|
| 48 |
+
"37": "TAX_ID",
|
| 49 |
+
"38": "TIME",
|
| 50 |
+
"39": "UNIQUE_IDENTIFIER",
|
| 51 |
+
"40": "URL",
|
| 52 |
+
"41": "USER_NAME",
|
| 53 |
+
"42": "VEHICLE_IDENTIFIER"
|
| 54 |
+
},
|
| 55 |
+
"initializer_range": 0.02,
|
| 56 |
+
"intermediate_size": 3072,
|
| 57 |
+
"label2id": {
|
| 58 |
+
"ACCOUNT_NUMBER": 1,
|
| 59 |
+
"ADDRESS": 2,
|
| 60 |
+
"API_KEY": 3,
|
| 61 |
+
"BANK_ROUTING_NUMBER": 4,
|
| 62 |
+
"BIOMETRIC_IDENTIFIER": 5,
|
| 63 |
+
"CERTIFICATE_LICENSE_NUMBER": 6,
|
| 64 |
+
"CITY": 7,
|
| 65 |
+
"COMPANY_NAME": 8,
|
| 66 |
+
"COORDINATE": 9,
|
| 67 |
+
"COUNTRY": 10,
|
| 68 |
+
"CREDIT_CARD_NUMBER": 11,
|
| 69 |
+
"CUSTOMER_ID": 12,
|
| 70 |
+
"CVV": 13,
|
| 71 |
+
"DATE": 14,
|
| 72 |
+
"DATE_OF_BIRTH": 15,
|
| 73 |
+
"DATE_TIME": 16,
|
| 74 |
+
"DEVICE_IDENTIFIER": 17,
|
| 75 |
+
"EMAIL": 18,
|
| 76 |
+
"EMPLOYEE_ID": 19,
|
| 77 |
+
"FIRST_NAME": 20,
|
| 78 |
+
"HEALTH_PLAN_BENEFICIARY_NUMBER": 21,
|
| 79 |
+
"IPV4": 22,
|
| 80 |
+
"IPV6": 23,
|
| 81 |
+
"LAST_NAME": 24,
|
| 82 |
+
"LICENSE_PLATE": 25,
|
| 83 |
+
"MEDICAL_RECORD_NUMBER": 26,
|
| 84 |
+
"NAME": 27,
|
| 85 |
+
"NATIONAL_ID": 28,
|
| 86 |
+
"O": 0,
|
| 87 |
+
"PASSWORD": 29,
|
| 88 |
+
"PHONE_NUMBER": 30,
|
| 89 |
+
"PIN": 31,
|
| 90 |
+
"POSTCODE": 32,
|
| 91 |
+
"SSN": 33,
|
| 92 |
+
"STATE": 34,
|
| 93 |
+
"STREET_ADDRESS": 35,
|
| 94 |
+
"SWIFT_BIC": 36,
|
| 95 |
+
"TAX_ID": 37,
|
| 96 |
+
"TIME": 38,
|
| 97 |
+
"UNIQUE_IDENTIFIER": 39,
|
| 98 |
+
"URL": 40,
|
| 99 |
+
"USER_NAME": 41,
|
| 100 |
+
"VEHICLE_IDENTIFIER": 42
|
| 101 |
+
},
|
| 102 |
+
"layer_norm_eps": 1e-07,
|
| 103 |
+
"legacy": true,
|
| 104 |
+
"max_position_embeddings": 512,
|
| 105 |
+
"max_relative_positions": -1,
|
| 106 |
+
"model_type": "deberta-v2",
|
| 107 |
+
"norm_rel_ebd": "layer_norm",
|
| 108 |
+
"num_attention_heads": 12,
|
| 109 |
+
"num_hidden_layers": 12,
|
| 110 |
+
"pad_token_id": 0,
|
| 111 |
+
"pooler_dropout": 0,
|
| 112 |
+
"pooler_hidden_act": "gelu",
|
| 113 |
+
"pooler_hidden_size": 768,
|
| 114 |
+
"pos_att_type": [
|
| 115 |
+
"p2c",
|
| 116 |
+
"c2p"
|
| 117 |
+
],
|
| 118 |
+
"position_biased_input": false,
|
| 119 |
+
"position_buckets": 256,
|
| 120 |
+
"relative_attention": true,
|
| 121 |
+
"share_att_key": true,
|
| 122 |
+
"transformers_version": "4.57.6",
|
| 123 |
+
"type_vocab_size": 0,
|
| 124 |
+
"vocab_size": 128100
|
| 125 |
+
}
|
gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab1968e81b8830f6578271899a3d997269f8fcf58579d6566068fb82ef657f41
|
| 3 |
+
size 741430532
|
gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66201f0bb6a842aaf1fa4d1f4feaaa101ca38f0e2841e511a78c16c379625b2c
|
| 3 |
+
size 1482984715
|
gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ccb9dee506e95ea6dd19b6ec3ce21d4d3cc3f1b01970439937ba6918dcb932d5
|
| 3 |
+
size 14645
|
gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:56f36e88329adcbc6a7f19a937d91abeb8dbfbfe389c74f13618f1d9852ac985
|
| 3 |
+
size 1465
|
gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/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 |
+
}
|
gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/spm.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
|
| 3 |
+
size 2464616
|
gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/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 |
+
}
|
gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/trainer_state.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 580,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 1.7245543415756182,
|
| 14 |
+
"grad_norm": 0.013835746794939041,
|
| 15 |
+
"learning_rate": 7.758620689655173e-07,
|
| 16 |
+
"loss": 0.0198,
|
| 17 |
+
"step": 500
|
| 18 |
+
}
|
| 19 |
+
],
|
| 20 |
+
"logging_steps": 500,
|
| 21 |
+
"max_steps": 580,
|
| 22 |
+
"num_input_tokens_seen": 0,
|
| 23 |
+
"num_train_epochs": 2,
|
| 24 |
+
"save_steps": 500,
|
| 25 |
+
"stateful_callbacks": {
|
| 26 |
+
"TrainerControl": {
|
| 27 |
+
"args": {
|
| 28 |
+
"should_epoch_stop": false,
|
| 29 |
+
"should_evaluate": false,
|
| 30 |
+
"should_log": false,
|
| 31 |
+
"should_save": true,
|
| 32 |
+
"should_training_stop": true
|
| 33 |
+
},
|
| 34 |
+
"attributes": {}
|
| 35 |
+
}
|
| 36 |
+
},
|
| 37 |
+
"total_flos": 2.443885042235328e+16,
|
| 38 |
+
"train_batch_size": 64,
|
| 39 |
+
"trial_name": null,
|
| 40 |
+
"trial_params": null
|
| 41 |
+
}
|
gretel-pii-ready-augmented/span-deberta-v3-base/checkpoint-580/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:badbe608928902060981a82397d4ac00e3c5954a33ef7cec1c7e357a5ba48b09
|
| 3 |
+
size 5905
|
gretel-pii-ready-augmented/span-deberta-v3-base/config.json
ADDED
|
@@ -0,0 +1,125 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "ADDRESS",
|
| 14 |
+
"3": "API_KEY",
|
| 15 |
+
"4": "BANK_ROUTING_NUMBER",
|
| 16 |
+
"5": "BIOMETRIC_IDENTIFIER",
|
| 17 |
+
"6": "CERTIFICATE_LICENSE_NUMBER",
|
| 18 |
+
"7": "CITY",
|
| 19 |
+
"8": "COMPANY_NAME",
|
| 20 |
+
"9": "COORDINATE",
|
| 21 |
+
"10": "COUNTRY",
|
| 22 |
+
"11": "CREDIT_CARD_NUMBER",
|
| 23 |
+
"12": "CUSTOMER_ID",
|
| 24 |
+
"13": "CVV",
|
| 25 |
+
"14": "DATE",
|
| 26 |
+
"15": "DATE_OF_BIRTH",
|
| 27 |
+
"16": "DATE_TIME",
|
| 28 |
+
"17": "DEVICE_IDENTIFIER",
|
| 29 |
+
"18": "EMAIL",
|
| 30 |
+
"19": "EMPLOYEE_ID",
|
| 31 |
+
"20": "FIRST_NAME",
|
| 32 |
+
"21": "HEALTH_PLAN_BENEFICIARY_NUMBER",
|
| 33 |
+
"22": "IPV4",
|
| 34 |
+
"23": "IPV6",
|
| 35 |
+
"24": "LAST_NAME",
|
| 36 |
+
"25": "LICENSE_PLATE",
|
| 37 |
+
"26": "MEDICAL_RECORD_NUMBER",
|
| 38 |
+
"27": "NAME",
|
| 39 |
+
"28": "NATIONAL_ID",
|
| 40 |
+
"29": "PASSWORD",
|
| 41 |
+
"30": "PHONE_NUMBER",
|
| 42 |
+
"31": "PIN",
|
| 43 |
+
"32": "POSTCODE",
|
| 44 |
+
"33": "SSN",
|
| 45 |
+
"34": "STATE",
|
| 46 |
+
"35": "STREET_ADDRESS",
|
| 47 |
+
"36": "SWIFT_BIC",
|
| 48 |
+
"37": "TAX_ID",
|
| 49 |
+
"38": "TIME",
|
| 50 |
+
"39": "UNIQUE_IDENTIFIER",
|
| 51 |
+
"40": "URL",
|
| 52 |
+
"41": "USER_NAME",
|
| 53 |
+
"42": "VEHICLE_IDENTIFIER"
|
| 54 |
+
},
|
| 55 |
+
"initializer_range": 0.02,
|
| 56 |
+
"intermediate_size": 3072,
|
| 57 |
+
"label2id": {
|
| 58 |
+
"ACCOUNT_NUMBER": 1,
|
| 59 |
+
"ADDRESS": 2,
|
| 60 |
+
"API_KEY": 3,
|
| 61 |
+
"BANK_ROUTING_NUMBER": 4,
|
| 62 |
+
"BIOMETRIC_IDENTIFIER": 5,
|
| 63 |
+
"CERTIFICATE_LICENSE_NUMBER": 6,
|
| 64 |
+
"CITY": 7,
|
| 65 |
+
"COMPANY_NAME": 8,
|
| 66 |
+
"COORDINATE": 9,
|
| 67 |
+
"COUNTRY": 10,
|
| 68 |
+
"CREDIT_CARD_NUMBER": 11,
|
| 69 |
+
"CUSTOMER_ID": 12,
|
| 70 |
+
"CVV": 13,
|
| 71 |
+
"DATE": 14,
|
| 72 |
+
"DATE_OF_BIRTH": 15,
|
| 73 |
+
"DATE_TIME": 16,
|
| 74 |
+
"DEVICE_IDENTIFIER": 17,
|
| 75 |
+
"EMAIL": 18,
|
| 76 |
+
"EMPLOYEE_ID": 19,
|
| 77 |
+
"FIRST_NAME": 20,
|
| 78 |
+
"HEALTH_PLAN_BENEFICIARY_NUMBER": 21,
|
| 79 |
+
"IPV4": 22,
|
| 80 |
+
"IPV6": 23,
|
| 81 |
+
"LAST_NAME": 24,
|
| 82 |
+
"LICENSE_PLATE": 25,
|
| 83 |
+
"MEDICAL_RECORD_NUMBER": 26,
|
| 84 |
+
"NAME": 27,
|
| 85 |
+
"NATIONAL_ID": 28,
|
| 86 |
+
"O": 0,
|
| 87 |
+
"PASSWORD": 29,
|
| 88 |
+
"PHONE_NUMBER": 30,
|
| 89 |
+
"PIN": 31,
|
| 90 |
+
"POSTCODE": 32,
|
| 91 |
+
"SSN": 33,
|
| 92 |
+
"STATE": 34,
|
| 93 |
+
"STREET_ADDRESS": 35,
|
| 94 |
+
"SWIFT_BIC": 36,
|
| 95 |
+
"TAX_ID": 37,
|
| 96 |
+
"TIME": 38,
|
| 97 |
+
"UNIQUE_IDENTIFIER": 39,
|
| 98 |
+
"URL": 40,
|
| 99 |
+
"USER_NAME": 41,
|
| 100 |
+
"VEHICLE_IDENTIFIER": 42
|
| 101 |
+
},
|
| 102 |
+
"layer_norm_eps": 1e-07,
|
| 103 |
+
"legacy": true,
|
| 104 |
+
"max_position_embeddings": 512,
|
| 105 |
+
"max_relative_positions": -1,
|
| 106 |
+
"model_type": "deberta-v2",
|
| 107 |
+
"norm_rel_ebd": "layer_norm",
|
| 108 |
+
"num_attention_heads": 12,
|
| 109 |
+
"num_hidden_layers": 12,
|
| 110 |
+
"pad_token_id": 0,
|
| 111 |
+
"pooler_dropout": 0,
|
| 112 |
+
"pooler_hidden_act": "gelu",
|
| 113 |
+
"pooler_hidden_size": 768,
|
| 114 |
+
"pos_att_type": [
|
| 115 |
+
"p2c",
|
| 116 |
+
"c2p"
|
| 117 |
+
],
|
| 118 |
+
"position_biased_input": false,
|
| 119 |
+
"position_buckets": 256,
|
| 120 |
+
"relative_attention": true,
|
| 121 |
+
"share_att_key": true,
|
| 122 |
+
"transformers_version": "4.57.6",
|
| 123 |
+
"type_vocab_size": 0,
|
| 124 |
+
"vocab_size": 128100
|
| 125 |
+
}
|
gretel-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:d633b84b80641f2b1aa914831a054386a770eb611ece98b6883d214ee18f3e8d
|
| 3 |
+
size 741492431
|
gretel-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 |
+
}
|
gretel-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
|
gretel-pii-ready-augmented/span-deberta-v3-base/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
gretel-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 |
+
}
|
gretel-pii-ready-augmented/span-deberta-v3-base/training_meta.json
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architecture": "span_based",
|
| 3 |
+
"base_model": "microsoft/deberta-v3-base",
|
| 4 |
+
"dataset": "gretel-pii-ready-augmented",
|
| 5 |
+
"label_list": [
|
| 6 |
+
"O",
|
| 7 |
+
"ACCOUNT_NUMBER",
|
| 8 |
+
"ADDRESS",
|
| 9 |
+
"API_KEY",
|
| 10 |
+
"BANK_ROUTING_NUMBER",
|
| 11 |
+
"BIOMETRIC_IDENTIFIER",
|
| 12 |
+
"CERTIFICATE_LICENSE_NUMBER",
|
| 13 |
+
"CITY",
|
| 14 |
+
"COMPANY_NAME",
|
| 15 |
+
"COORDINATE",
|
| 16 |
+
"COUNTRY",
|
| 17 |
+
"CREDIT_CARD_NUMBER",
|
| 18 |
+
"CUSTOMER_ID",
|
| 19 |
+
"CVV",
|
| 20 |
+
"DATE",
|
| 21 |
+
"DATE_OF_BIRTH",
|
| 22 |
+
"DATE_TIME",
|
| 23 |
+
"DEVICE_IDENTIFIER",
|
| 24 |
+
"EMAIL",
|
| 25 |
+
"EMPLOYEE_ID",
|
| 26 |
+
"FIRST_NAME",
|
| 27 |
+
"HEALTH_PLAN_BENEFICIARY_NUMBER",
|
| 28 |
+
"IPV4",
|
| 29 |
+
"IPV6",
|
| 30 |
+
"LAST_NAME",
|
| 31 |
+
"LICENSE_PLATE",
|
| 32 |
+
"MEDICAL_RECORD_NUMBER",
|
| 33 |
+
"NAME",
|
| 34 |
+
"NATIONAL_ID",
|
| 35 |
+
"PASSWORD",
|
| 36 |
+
"PHONE_NUMBER",
|
| 37 |
+
"PIN",
|
| 38 |
+
"POSTCODE",
|
| 39 |
+
"SSN",
|
| 40 |
+
"STATE",
|
| 41 |
+
"STREET_ADDRESS",
|
| 42 |
+
"SWIFT_BIC",
|
| 43 |
+
"TAX_ID",
|
| 44 |
+
"TIME",
|
| 45 |
+
"UNIQUE_IDENTIFIER",
|
| 46 |
+
"URL",
|
| 47 |
+
"USER_NAME",
|
| 48 |
+
"VEHICLE_IDENTIFIER"
|
| 49 |
+
],
|
| 50 |
+
"max_seq_length": 512,
|
| 51 |
+
"max_span_width": 30,
|
| 52 |
+
"timestamp": "2026-04-07T03:34:15.459649"
|
| 53 |
+
}
|