Upload 12 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+ai4privacy-mdeberta-v3-base-general-preprocessed/tokenizer.json filter=lfs diff=lfs merge=lfs -text

ai4privacy-mdeberta-v3-base-general-preprocessed/added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "[MASK]": 250101
+}

ai4privacy-mdeberta-v3-base-general-preprocessed/config.json

@@ -0,0 +1,109 @@
+{
+  "_name_or_path": "microsoft/mdeberta-v3-base",
+  "architectures": [
+    "DebertaV2ForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "B-ACCOUNTNUM",
+    "1": "B-BUILDINGNUM",
+    "2": "B-CITY",
+    "3": "B-CREDITCARDNUMBER",
+    "4": "B-DATEOFBIRTH",
+    "5": "B-DRIVERLICENSENUM",
+    "6": "B-EMAIL",
+    "7": "B-GIVENNAME",
+    "8": "B-IDCARDNUM",
+    "9": "B-PASSWORD",
+    "10": "B-SOCIALNUM",
+    "11": "B-STREET",
+    "12": "B-SURNAME",
+    "13": "B-TAXNUM",
+    "14": "B-TELEPHONENUM",
+    "15": "B-USERNAME",
+    "16": "B-ZIPCODE",
+    "17": "I-ACCOUNTNUM",
+    "18": "I-BUILDINGNUM",
+    "19": "I-CITY",
+    "20": "I-CREDITCARDNUMBER",
+    "21": "I-DATEOFBIRTH",
+    "22": "I-DRIVERLICENSENUM",
+    "23": "I-EMAIL",
+    "24": "I-GIVENNAME",
+    "25": "I-IDCARDNUM",
+    "26": "I-PASSWORD",
+    "27": "I-SOCIALNUM",
+    "28": "I-STREET",
+    "29": "I-SURNAME",
+    "30": "I-TAXNUM",
+    "31": "I-TELEPHONENUM",
+    "32": "I-USERNAME",
+    "33": "I-ZIPCODE",
+    "34": "O"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "B-ACCOUNTNUM": 0,
+    "B-BUILDINGNUM": 1,
+    "B-CITY": 2,
+    "B-CREDITCARDNUMBER": 3,
+    "B-DATEOFBIRTH": 4,
+    "B-DRIVERLICENSENUM": 5,
+    "B-EMAIL": 6,
+    "B-GIVENNAME": 7,
+    "B-IDCARDNUM": 8,
+    "B-PASSWORD": 9,
+    "B-SOCIALNUM": 10,
+    "B-STREET": 11,
+    "B-SURNAME": 12,
+    "B-TAXNUM": 13,
+    "B-TELEPHONENUM": 14,
+    "B-USERNAME": 15,
+    "B-ZIPCODE": 16,
+    "I-ACCOUNTNUM": 17,
+    "I-BUILDINGNUM": 18,
+    "I-CITY": 19,
+    "I-CREDITCARDNUMBER": 20,
+    "I-DATEOFBIRTH": 21,
+    "I-DRIVERLICENSENUM": 22,
+    "I-EMAIL": 23,
+    "I-GIVENNAME": 24,
+    "I-IDCARDNUM": 25,
+    "I-PASSWORD": 26,
+    "I-SOCIALNUM": 27,
+    "I-STREET": 28,
+    "I-SURNAME": 29,
+    "I-TAXNUM": 30,
+    "I-TELEPHONENUM": 31,
+    "I-USERNAME": 32,
+    "I-ZIPCODE": 33,
+    "O": 34
+  },
+  "layer_norm_eps": 1e-07,
+  "max_position_embeddings": 512,
+  "max_relative_positions": -1,
+  "model_type": "deberta-v2",
+  "norm_rel_ebd": "layer_norm",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 768,
+  "pos_att_type": [
+    "p2c",
+    "c2p"
+  ],
+  "position_biased_input": false,
+  "position_buckets": 256,
+  "relative_attention": true,
+  "share_att_key": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.46.0",
+  "type_vocab_size": 0,
+  "vocab_size": 251000
+}

ai4privacy-mdeberta-v3-base-general-preprocessed/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:160256ad0460e09ab6edf3ef8c7f38995acae6a7efa9069073e96b8ac081d9aa
+size 1113007172

ai4privacy-mdeberta-v3-base-general-preprocessed/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70bf8620b7ecb98f0bd016096bc5f7d22181cefdecf9ecc37379678931111401
+size 2226132602

ai4privacy-mdeberta-v3-base-general-preprocessed/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9299a7ff31388e4bac3d48956b48e54b0eedc5c1f830dae9cb6c25260f05ac30
+size 14244

ai4privacy-mdeberta-v3-base-general-preprocessed/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e148c218bcd0b0a4869b7ca0f771dcc4651965546b179da5e6b22a519d98cd45
+size 1064

ai4privacy-mdeberta-v3-base-general-preprocessed/special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "[CLS]",
+  "cls_token": "[CLS]",
+  "eos_token": "[SEP]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

ai4privacy-mdeberta-v3-base-general-preprocessed/spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13c8d666d62a7bc4ac8f040aab68e942c861f93303156cc28f5c7e885d86d6e3
+size 4305025

ai4privacy-mdeberta-v3-base-general-preprocessed/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ddbe35b768b22d6cd0d61a0a88ea3a188b7c00bbb7b825f9f247cf0b79c2365
+size 16350764

ai4privacy-mdeberta-v3-base-general-preprocessed/tokenizer_config.json

@@ -0,0 +1,860 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "▁<extra_id_99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250002": {
+      "content": "▁<extra_id_98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250003": {
+      "content": "▁<extra_id_97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250004": {
+      "content": "▁<extra_id_96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250005": {
+      "content": "▁<extra_id_95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250006": {
+      "content": "▁<extra_id_94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250007": {
+      "content": "▁<extra_id_93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250008": {
+      "content": "▁<extra_id_92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250009": {
+      "content": "▁<extra_id_91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250010": {
+      "content": "▁<extra_id_90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250011": {
+      "content": "▁<extra_id_89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250012": {
+      "content": "▁<extra_id_88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250013": {
+      "content": "▁<extra_id_87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250014": {
+      "content": "▁<extra_id_86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250015": {
+      "content": "▁<extra_id_85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250016": {
+      "content": "▁<extra_id_84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250017": {
+      "content": "▁<extra_id_83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250018": {
+      "content": "▁<extra_id_82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250019": {
+      "content": "▁<extra_id_81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250020": {
+      "content": "▁<extra_id_80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250021": {
+      "content": "▁<extra_id_79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250022": {
+      "content": "▁<extra_id_78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250023": {
+      "content": "▁<extra_id_77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250024": {
+      "content": "▁<extra_id_76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250025": {
+      "content": "▁<extra_id_75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250026": {
+      "content": "▁<extra_id_74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250027": {
+      "content": "▁<extra_id_73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250028": {
+      "content": "▁<extra_id_72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250029": {
+      "content": "▁<extra_id_71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250030": {
+      "content": "▁<extra_id_70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250031": {
+      "content": "▁<extra_id_69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250032": {
+      "content": "▁<extra_id_68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250033": {
+      "content": "▁<extra_id_67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250034": {
+      "content": "▁<extra_id_66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250035": {
+      "content": "▁<extra_id_65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250036": {
+      "content": "▁<extra_id_64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250037": {
+      "content": "▁<extra_id_63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250038": {
+      "content": "▁<extra_id_62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250039": {
+      "content": "▁<extra_id_61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250040": {
+      "content": "▁<extra_id_60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250041": {
+      "content": "▁<extra_id_59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250042": {
+      "content": "▁<extra_id_58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250043": {
+      "content": "▁<extra_id_57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250044": {
+      "content": "▁<extra_id_56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250045": {
+      "content": "▁<extra_id_55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250046": {
+      "content": "▁<extra_id_54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250047": {
+      "content": "▁<extra_id_53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250048": {
+      "content": "▁<extra_id_52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250049": {
+      "content": "▁<extra_id_51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250050": {
+      "content": "▁<extra_id_50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250051": {
+      "content": "▁<extra_id_49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250052": {
+      "content": "▁<extra_id_48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250053": {
+      "content": "▁<extra_id_47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250054": {
+      "content": "▁<extra_id_46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250055": {
+      "content": "▁<extra_id_45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250056": {
+      "content": "▁<extra_id_44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250057": {
+      "content": "▁<extra_id_43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250058": {
+      "content": "▁<extra_id_42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250059": {
+      "content": "▁<extra_id_41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250060": {
+      "content": "▁<extra_id_40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250061": {
+      "content": "▁<extra_id_39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250062": {
+      "content": "▁<extra_id_38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250063": {
+      "content": "▁<extra_id_37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250064": {
+      "content": "▁<extra_id_36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250065": {
+      "content": "▁<extra_id_35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250066": {
+      "content": "▁<extra_id_34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250067": {
+      "content": "▁<extra_id_33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250068": {
+      "content": "▁<extra_id_32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250069": {
+      "content": "▁<extra_id_31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250070": {
+      "content": "▁<extra_id_30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250071": {
+      "content": "▁<extra_id_29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250072": {
+      "content": "▁<extra_id_28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250073": {
+      "content": "▁<extra_id_27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250074": {
+      "content": "▁<extra_id_26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250075": {
+      "content": "▁<extra_id_25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250076": {
+      "content": "▁<extra_id_24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250077": {
+      "content": "▁<extra_id_23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250078": {
+      "content": "▁<extra_id_22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250079": {
+      "content": "▁<extra_id_21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250080": {
+      "content": "▁<extra_id_20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250081": {
+      "content": "▁<extra_id_19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250082": {
+      "content": "▁<extra_id_18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250083": {
+      "content": "▁<extra_id_17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250084": {
+      "content": "▁<extra_id_16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250085": {
+      "content": "▁<extra_id_15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250086": {
+      "content": "▁<extra_id_14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250087": {
+      "content": "▁<extra_id_13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250088": {
+      "content": "▁<extra_id_12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250089": {
+      "content": "▁<extra_id_11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250090": {
+      "content": "▁<extra_id_10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250091": {
+      "content": "▁<extra_id_9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250092": {
+      "content": "▁<extra_id_8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250093": {
+      "content": "▁<extra_id_7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250094": {
+      "content": "▁<extra_id_6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250095": {
+      "content": "▁<extra_id_5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250096": {
+      "content": "▁<extra_id_4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250097": {
+      "content": "▁<extra_id_3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250098": {
+      "content": "▁<extra_id_2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250099": {
+      "content": "▁<extra_id_1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250100": {
+      "content": "▁<extra_id_0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250101": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "[CLS]",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "[CLS]",
+  "do_lower_case": false,
+  "eos_token": "[SEP]",
+  "mask_token": "[MASK]",
+  "max_length": 512,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "sp_model_kwargs": {},
+  "split_by_punct": false,
+  "tokenizer_class": "DebertaV2Tokenizer",
+  "truncation": true,
+  "unk_token": "[UNK]",
+  "vocab_type": "spm"
+}

ai4privacy-mdeberta-v3-base-general-preprocessed/trainer_state.json

@@ -0,0 +1,300 @@
+{
+  "best_metric": 0.01948946714401245,
+  "best_model_checkpoint": "./results-gen-3/checkpoint-16839",
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 16839,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.08907892392659897,
+      "grad_norm": 2.219806432723999,
+      "learning_rate": 2.910921076073401e-05,
+      "loss": 0.2527,
+      "step": 500
+    },
+    {
+      "epoch": 0.17815784785319794,
+      "grad_norm": 0.8369982242584229,
+      "learning_rate": 2.8218421521468022e-05,
+      "loss": 0.0603,
+      "step": 1000
+    },
+    {
+      "epoch": 0.2672367717797969,
+      "grad_norm": 2.066608190536499,
+      "learning_rate": 2.732763228220203e-05,
+      "loss": 0.0479,
+      "step": 1500
+    },
+    {
+      "epoch": 0.3563156957063959,
+      "grad_norm": 0.8151403069496155,
+      "learning_rate": 2.643684304293604e-05,
+      "loss": 0.0416,
+      "step": 2000
+    },
+    {
+      "epoch": 0.4453946196329948,
+      "grad_norm": 0.5260671973228455,
+      "learning_rate": 2.554605380367005e-05,
+      "loss": 0.0375,
+      "step": 2500
+    },
+    {
+      "epoch": 0.5344735435595938,
+      "grad_norm": 0.7381831407546997,
+      "learning_rate": 2.4655264564404063e-05,
+      "loss": 0.0362,
+      "step": 3000
+    },
+    {
+      "epoch": 0.6235524674861928,
+      "grad_norm": 0.6741832494735718,
+      "learning_rate": 2.3764475325138073e-05,
+      "loss": 0.0332,
+      "step": 3500
+    },
+    {
+      "epoch": 0.7126313914127917,
+      "grad_norm": 1.0195770263671875,
+      "learning_rate": 2.2873686085872084e-05,
+      "loss": 0.0325,
+      "step": 4000
+    },
+    {
+      "epoch": 0.8017103153393907,
+      "grad_norm": 0.6665781736373901,
+      "learning_rate": 2.1982896846606095e-05,
+      "loss": 0.032,
+      "step": 4500
+    },
+    {
+      "epoch": 0.8907892392659896,
+      "grad_norm": 0.7846745252609253,
+      "learning_rate": 2.1092107607340103e-05,
+      "loss": 0.0306,
+      "step": 5000
+    },
+    {
+      "epoch": 0.9798681631925886,
+      "grad_norm": 0.32996097207069397,
+      "learning_rate": 2.0201318368074113e-05,
+      "loss": 0.029,
+      "step": 5500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9906996625111044,
+      "eval_f1": 0.9087714540750031,
+      "eval_loss": 0.025583524256944656,
+      "eval_precision": 0.8997740188839666,
+      "eval_recall": 0.9179506494099386,
+      "eval_runtime": 144.0373,
+      "eval_samples_per_second": 564.986,
+      "eval_steps_per_second": 9.747,
+      "step": 5613
+    },
+    {
+      "epoch": 1.0689470871191875,
+      "grad_norm": 0.21343424916267395,
+      "learning_rate": 1.9310529128808124e-05,
+      "loss": 0.0252,
+      "step": 6000
+    },
+    {
+      "epoch": 1.1580260110457865,
+      "grad_norm": 0.483236700296402,
+      "learning_rate": 1.8419739889542135e-05,
+      "loss": 0.0237,
+      "step": 6500
+    },
+    {
+      "epoch": 1.2471049349723855,
+      "grad_norm": 0.4517042934894562,
+      "learning_rate": 1.7528950650276143e-05,
+      "loss": 0.0232,
+      "step": 7000
+    },
+    {
+      "epoch": 1.3361838588989845,
+      "grad_norm": 0.4084126651287079,
+      "learning_rate": 1.6638161411010153e-05,
+      "loss": 0.0227,
+      "step": 7500
+    },
+    {
+      "epoch": 1.4252627828255835,
+      "grad_norm": 0.9308103322982788,
+      "learning_rate": 1.5747372171744164e-05,
+      "loss": 0.0235,
+      "step": 8000
+    },
+    {
+      "epoch": 1.5143417067521825,
+      "grad_norm": 0.4621828496456146,
+      "learning_rate": 1.4856582932478175e-05,
+      "loss": 0.0231,
+      "step": 8500
+    },
+    {
+      "epoch": 1.6034206306787815,
+      "grad_norm": 0.41997483372688293,
+      "learning_rate": 1.3965793693212186e-05,
+      "loss": 0.0219,
+      "step": 9000
+    },
+    {
+      "epoch": 1.6924995546053805,
+      "grad_norm": 0.5039160847663879,
+      "learning_rate": 1.3075004453946195e-05,
+      "loss": 0.0208,
+      "step": 9500
+    },
+    {
+      "epoch": 1.7815784785319795,
+      "grad_norm": 0.20699962973594666,
+      "learning_rate": 1.2184215214680208e-05,
+      "loss": 0.0217,
+      "step": 10000
+    },
+    {
+      "epoch": 1.8706574024585783,
+      "grad_norm": 0.449458509683609,
+      "learning_rate": 1.1293425975414217e-05,
+      "loss": 0.0212,
+      "step": 10500
+    },
+    {
+      "epoch": 1.9597363263851773,
+      "grad_norm": 0.1629885584115982,
+      "learning_rate": 1.0402636736148228e-05,
+      "loss": 0.0209,
+      "step": 11000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9922627801604135,
+      "eval_f1": 0.9241052128603608,
+      "eval_loss": 0.02089638262987137,
+      "eval_precision": 0.921170813581619,
+      "eval_recall": 0.9270583670022141,
+      "eval_runtime": 127.6731,
+      "eval_samples_per_second": 637.402,
+      "eval_steps_per_second": 10.997,
+      "step": 11226
+    },
+    {
+      "epoch": 2.048815250311776,
+      "grad_norm": 0.37108752131462097,
+      "learning_rate": 9.511847496882237e-06,
+      "loss": 0.018,
+      "step": 11500
+    },
+    {
+      "epoch": 2.137894174238375,
+      "grad_norm": 0.6335046291351318,
+      "learning_rate": 8.621058257616248e-06,
+      "loss": 0.0167,
+      "step": 12000
+    },
+    {
+      "epoch": 2.226973098164974,
+      "grad_norm": 1.3230624198913574,
+      "learning_rate": 7.730269018350257e-06,
+      "loss": 0.0166,
+      "step": 12500
+    },
+    {
+      "epoch": 2.316052022091573,
+      "grad_norm": 0.6974636912345886,
+      "learning_rate": 6.839479779084269e-06,
+      "loss": 0.0167,
+      "step": 13000
+    },
+    {
+      "epoch": 2.405130946018172,
+      "grad_norm": 0.25162258744239807,
+      "learning_rate": 5.9486905398182796e-06,
+      "loss": 0.0158,
+      "step": 13500
+    },
+    {
+      "epoch": 2.494209869944771,
+      "grad_norm": 0.34794881939888,
+      "learning_rate": 5.0579013005522896e-06,
+      "loss": 0.0157,
+      "step": 14000
+    },
+    {
+      "epoch": 2.58328879387137,
+      "grad_norm": 0.20756091177463531,
+      "learning_rate": 4.1671120612863e-06,
+      "loss": 0.0156,
+      "step": 14500
+    },
+    {
+      "epoch": 2.672367717797969,
+      "grad_norm": 0.567460298538208,
+      "learning_rate": 3.27632282202031e-06,
+      "loss": 0.0153,
+      "step": 15000
+    },
+    {
+      "epoch": 2.761446641724568,
+      "grad_norm": 0.4445217549800873,
+      "learning_rate": 2.3855335827543205e-06,
+      "loss": 0.0154,
+      "step": 15500
+    },
+    {
+      "epoch": 2.850525565651167,
+      "grad_norm": 0.19330230355262756,
+      "learning_rate": 1.4947443434883307e-06,
+      "loss": 0.0154,
+      "step": 16000
+    },
+    {
+      "epoch": 2.939604489577766,
+      "grad_norm": 0.5202648043632507,
+      "learning_rate": 6.03955104222341e-07,
+      "loss": 0.0148,
+      "step": 16500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9930438076738344,
+      "eval_f1": 0.931667281969727,
+      "eval_loss": 0.01948946714401245,
+      "eval_precision": 0.9273010955713591,
+      "eval_recall": 0.9360747791549682,
+      "eval_runtime": 110.8205,
+      "eval_samples_per_second": 734.331,
+      "eval_steps_per_second": 12.669,
+      "step": 16839
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 16839,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.600668859059613e+16,
+  "train_batch_size": 58,
+  "trial_name": null,
+  "trial_params": null
+}

ai4privacy-mdeberta-v3-base-general-preprocessed/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26b2a090130a8b1eaf88938c667db49e178adaf0e3073a3f5dead269183f4938
+size 5240