PuxAI commited on
Commit
f1f1020
·
verified ·
1 Parent(s): 3f3ee94

Upload folder using huggingface_hub

Browse files
Files changed (32) hide show
  1. gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/config.json +76 -0
  2. gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/model.safetensors +3 -0
  3. gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/optimizer.pt +3 -0
  4. gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/rng_state.pth +3 -0
  5. gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/scheduler.pt +3 -0
  6. gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/tokenizer.json +0 -0
  7. gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/tokenizer_config.json +15 -0
  8. gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/trainer_state.json +72 -0
  9. gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/training_args.bin +3 -0
  10. gorkem-pii-ready/TokenBased-BERT/checkpoint-362/config.json +76 -0
  11. gorkem-pii-ready/TokenBased-BERT/checkpoint-362/model.safetensors +3 -0
  12. gorkem-pii-ready/TokenBased-BERT/checkpoint-362/optimizer.pt +3 -0
  13. gorkem-pii-ready/TokenBased-BERT/checkpoint-362/rng_state.pth +3 -0
  14. gorkem-pii-ready/TokenBased-BERT/checkpoint-362/scheduler.pt +3 -0
  15. gorkem-pii-ready/TokenBased-BERT/checkpoint-362/tokenizer.json +0 -0
  16. gorkem-pii-ready/TokenBased-BERT/checkpoint-362/tokenizer_config.json +15 -0
  17. gorkem-pii-ready/TokenBased-BERT/checkpoint-362/trainer_state.json +42 -0
  18. gorkem-pii-ready/TokenBased-BERT/checkpoint-362/training_args.bin +3 -0
  19. gorkem-pii-ready/TokenBased-BERT/checkpoint-724/config.json +76 -0
  20. gorkem-pii-ready/TokenBased-BERT/checkpoint-724/model.safetensors +3 -0
  21. gorkem-pii-ready/TokenBased-BERT/checkpoint-724/optimizer.pt +3 -0
  22. gorkem-pii-ready/TokenBased-BERT/checkpoint-724/rng_state.pth +3 -0
  23. gorkem-pii-ready/TokenBased-BERT/checkpoint-724/scheduler.pt +3 -0
  24. gorkem-pii-ready/TokenBased-BERT/checkpoint-724/tokenizer.json +0 -0
  25. gorkem-pii-ready/TokenBased-BERT/checkpoint-724/tokenizer_config.json +15 -0
  26. gorkem-pii-ready/TokenBased-BERT/checkpoint-724/trainer_state.json +57 -0
  27. gorkem-pii-ready/TokenBased-BERT/checkpoint-724/training_args.bin +3 -0
  28. gorkem-pii-ready/TokenBased-BERT/config.json +76 -0
  29. gorkem-pii-ready/TokenBased-BERT/model.safetensors +3 -0
  30. gorkem-pii-ready/TokenBased-BERT/tokenizer.json +0 -0
  31. gorkem-pii-ready/TokenBased-BERT/tokenizer_config.json +15 -0
  32. gorkem-pii-ready/TokenBased-BERT/training_args.bin +3 -0
gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/config.json ADDED
@@ -0,0 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "BertForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": null,
8
+ "classifier_dropout": null,
9
+ "directionality": "bidi",
10
+ "dtype": "float32",
11
+ "eos_token_id": null,
12
+ "hidden_act": "gelu",
13
+ "hidden_dropout_prob": 0.1,
14
+ "hidden_size": 768,
15
+ "id2label": {
16
+ "0": "O",
17
+ "1": "B-ADDRESS",
18
+ "2": "I-ADDRESS",
19
+ "3": "B-CREDIT_CARD",
20
+ "4": "I-CREDIT_CARD",
21
+ "5": "B-CRYPTO_ADDRESS",
22
+ "6": "I-CRYPTO_ADDRESS",
23
+ "7": "B-EMAIL",
24
+ "8": "I-EMAIL",
25
+ "9": "B-IBAN",
26
+ "10": "I-IBAN",
27
+ "11": "B-OFF_PLATFORM_ATTEMPT",
28
+ "12": "I-OFF_PLATFORM_ATTEMPT",
29
+ "13": "B-PHONE",
30
+ "14": "I-PHONE",
31
+ "15": "B-SOCIAL_MEDIA",
32
+ "16": "I-SOCIAL_MEDIA",
33
+ "17": "B-URL",
34
+ "18": "I-URL"
35
+ },
36
+ "initializer_range": 0.02,
37
+ "intermediate_size": 3072,
38
+ "is_decoder": false,
39
+ "label2id": {
40
+ "B-ADDRESS": 1,
41
+ "B-CREDIT_CARD": 3,
42
+ "B-CRYPTO_ADDRESS": 5,
43
+ "B-EMAIL": 7,
44
+ "B-IBAN": 9,
45
+ "B-OFF_PLATFORM_ATTEMPT": 11,
46
+ "B-PHONE": 13,
47
+ "B-SOCIAL_MEDIA": 15,
48
+ "B-URL": 17,
49
+ "I-ADDRESS": 2,
50
+ "I-CREDIT_CARD": 4,
51
+ "I-CRYPTO_ADDRESS": 6,
52
+ "I-EMAIL": 8,
53
+ "I-IBAN": 10,
54
+ "I-OFF_PLATFORM_ATTEMPT": 12,
55
+ "I-PHONE": 14,
56
+ "I-SOCIAL_MEDIA": 16,
57
+ "I-URL": 18,
58
+ "O": 0
59
+ },
60
+ "layer_norm_eps": 1e-12,
61
+ "max_position_embeddings": 512,
62
+ "model_type": "bert",
63
+ "num_attention_heads": 12,
64
+ "num_hidden_layers": 12,
65
+ "pad_token_id": 0,
66
+ "pooler_fc_size": 768,
67
+ "pooler_num_attention_heads": 12,
68
+ "pooler_num_fc_layers": 3,
69
+ "pooler_size_per_head": 128,
70
+ "pooler_type": "first_token_transform",
71
+ "tie_word_embeddings": true,
72
+ "transformers_version": "5.3.0",
73
+ "type_vocab_size": 2,
74
+ "use_cache": false,
75
+ "vocab_size": 119547
76
+ }
gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97f5839fe9e4c978611fb55c0a0c3bc06b9faebb1c9e37a05c86bf7ed2ef729b
3
+ size 709133172
gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d734fe06444994316760235eb88afc61d4d63373564df2b3a66e973ec8c3d56a
3
+ size 1418389387
gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8916ff84cfc6edb5be04ac53a2ddf8ebf96734d83271b966b6c8afaa7acb8691
3
+ size 14645
gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd6d489d96b568ea8a85c6e12c27dd8c07df8a55283c6c6562e85604ad29bf37
3
+ size 1465
gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "cls_token": "[CLS]",
5
+ "do_lower_case": false,
6
+ "is_local": false,
7
+ "mask_token": "[MASK]",
8
+ "model_max_length": 512,
9
+ "pad_token": "[PAD]",
10
+ "sep_token": "[SEP]",
11
+ "strip_accents": null,
12
+ "tokenize_chinese_chars": true,
13
+ "tokenizer_class": "BertTokenizer",
14
+ "unk_token": "[UNK]"
15
+ }
gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/trainer_state.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1086,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "eval_loss": 0.023628847673535347,
15
+ "eval_runtime": 7.6533,
16
+ "eval_samples_per_second": 265.376,
17
+ "eval_steps_per_second": 33.188,
18
+ "step": 362
19
+ },
20
+ {
21
+ "epoch": 1.3812154696132597,
22
+ "grad_norm": 0.11147041618824005,
23
+ "learning_rate": 1.0810313075506448e-05,
24
+ "loss": 0.048348953247070316,
25
+ "step": 500
26
+ },
27
+ {
28
+ "epoch": 2.0,
29
+ "eval_loss": 0.01673239655792713,
30
+ "eval_runtime": 7.3819,
31
+ "eval_samples_per_second": 275.134,
32
+ "eval_steps_per_second": 34.409,
33
+ "step": 724
34
+ },
35
+ {
36
+ "epoch": 2.7624309392265194,
37
+ "grad_norm": 0.29888421297073364,
38
+ "learning_rate": 1.6022099447513815e-06,
39
+ "loss": 0.009993391990661621,
40
+ "step": 1000
41
+ },
42
+ {
43
+ "epoch": 3.0,
44
+ "eval_loss": 0.015370420180261135,
45
+ "eval_runtime": 7.0685,
46
+ "eval_samples_per_second": 287.33,
47
+ "eval_steps_per_second": 35.934,
48
+ "step": 1086
49
+ }
50
+ ],
51
+ "logging_steps": 500,
52
+ "max_steps": 1086,
53
+ "num_input_tokens_seen": 0,
54
+ "num_train_epochs": 3,
55
+ "save_steps": 500,
56
+ "stateful_callbacks": {
57
+ "TrainerControl": {
58
+ "args": {
59
+ "should_epoch_stop": false,
60
+ "should_evaluate": false,
61
+ "should_log": false,
62
+ "should_save": true,
63
+ "should_training_stop": true
64
+ },
65
+ "attributes": {}
66
+ }
67
+ },
68
+ "total_flos": 9068652356539392.0,
69
+ "train_batch_size": 32,
70
+ "trial_name": null,
71
+ "trial_params": null
72
+ }
gorkem-pii-ready/TokenBased-BERT/checkpoint-1086/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c89573ad34be37ec1bd0425d17a9ec938dcb4bb1682cdabd338ef2e2452dff4
3
+ size 5201
gorkem-pii-ready/TokenBased-BERT/checkpoint-362/config.json ADDED
@@ -0,0 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "BertForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": null,
8
+ "classifier_dropout": null,
9
+ "directionality": "bidi",
10
+ "dtype": "float32",
11
+ "eos_token_id": null,
12
+ "hidden_act": "gelu",
13
+ "hidden_dropout_prob": 0.1,
14
+ "hidden_size": 768,
15
+ "id2label": {
16
+ "0": "O",
17
+ "1": "B-ADDRESS",
18
+ "2": "I-ADDRESS",
19
+ "3": "B-CREDIT_CARD",
20
+ "4": "I-CREDIT_CARD",
21
+ "5": "B-CRYPTO_ADDRESS",
22
+ "6": "I-CRYPTO_ADDRESS",
23
+ "7": "B-EMAIL",
24
+ "8": "I-EMAIL",
25
+ "9": "B-IBAN",
26
+ "10": "I-IBAN",
27
+ "11": "B-OFF_PLATFORM_ATTEMPT",
28
+ "12": "I-OFF_PLATFORM_ATTEMPT",
29
+ "13": "B-PHONE",
30
+ "14": "I-PHONE",
31
+ "15": "B-SOCIAL_MEDIA",
32
+ "16": "I-SOCIAL_MEDIA",
33
+ "17": "B-URL",
34
+ "18": "I-URL"
35
+ },
36
+ "initializer_range": 0.02,
37
+ "intermediate_size": 3072,
38
+ "is_decoder": false,
39
+ "label2id": {
40
+ "B-ADDRESS": 1,
41
+ "B-CREDIT_CARD": 3,
42
+ "B-CRYPTO_ADDRESS": 5,
43
+ "B-EMAIL": 7,
44
+ "B-IBAN": 9,
45
+ "B-OFF_PLATFORM_ATTEMPT": 11,
46
+ "B-PHONE": 13,
47
+ "B-SOCIAL_MEDIA": 15,
48
+ "B-URL": 17,
49
+ "I-ADDRESS": 2,
50
+ "I-CREDIT_CARD": 4,
51
+ "I-CRYPTO_ADDRESS": 6,
52
+ "I-EMAIL": 8,
53
+ "I-IBAN": 10,
54
+ "I-OFF_PLATFORM_ATTEMPT": 12,
55
+ "I-PHONE": 14,
56
+ "I-SOCIAL_MEDIA": 16,
57
+ "I-URL": 18,
58
+ "O": 0
59
+ },
60
+ "layer_norm_eps": 1e-12,
61
+ "max_position_embeddings": 512,
62
+ "model_type": "bert",
63
+ "num_attention_heads": 12,
64
+ "num_hidden_layers": 12,
65
+ "pad_token_id": 0,
66
+ "pooler_fc_size": 768,
67
+ "pooler_num_attention_heads": 12,
68
+ "pooler_num_fc_layers": 3,
69
+ "pooler_size_per_head": 128,
70
+ "pooler_type": "first_token_transform",
71
+ "tie_word_embeddings": true,
72
+ "transformers_version": "5.3.0",
73
+ "type_vocab_size": 2,
74
+ "use_cache": false,
75
+ "vocab_size": 119547
76
+ }
gorkem-pii-ready/TokenBased-BERT/checkpoint-362/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6504103b1f0ebfe8609b92635e66dfd1721a486c995a515f111ff0ed97a89321
3
+ size 709133172
gorkem-pii-ready/TokenBased-BERT/checkpoint-362/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b30ac2bf6ed9803506f0e07afbbe4b979549b4b701533b1d38b87565b6d548b3
3
+ size 1418389387
gorkem-pii-ready/TokenBased-BERT/checkpoint-362/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43ccf712be76ebc2b2c4c98c94185c9f96e9c3f04d70480f07780f87fc1e7570
3
+ size 14645
gorkem-pii-ready/TokenBased-BERT/checkpoint-362/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2938c937e5c5f20bdb270519026fa116e12228bddf519437daa253056363dc71
3
+ size 1465
gorkem-pii-ready/TokenBased-BERT/checkpoint-362/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
gorkem-pii-ready/TokenBased-BERT/checkpoint-362/tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "cls_token": "[CLS]",
5
+ "do_lower_case": false,
6
+ "is_local": false,
7
+ "mask_token": "[MASK]",
8
+ "model_max_length": 512,
9
+ "pad_token": "[PAD]",
10
+ "sep_token": "[SEP]",
11
+ "strip_accents": null,
12
+ "tokenize_chinese_chars": true,
13
+ "tokenizer_class": "BertTokenizer",
14
+ "unk_token": "[UNK]"
15
+ }
gorkem-pii-ready/TokenBased-BERT/checkpoint-362/trainer_state.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 362,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "eval_loss": 0.023628847673535347,
15
+ "eval_runtime": 7.6533,
16
+ "eval_samples_per_second": 265.376,
17
+ "eval_steps_per_second": 33.188,
18
+ "step": 362
19
+ }
20
+ ],
21
+ "logging_steps": 500,
22
+ "max_steps": 1086,
23
+ "num_input_tokens_seen": 0,
24
+ "num_train_epochs": 3,
25
+ "save_steps": 500,
26
+ "stateful_callbacks": {
27
+ "TrainerControl": {
28
+ "args": {
29
+ "should_epoch_stop": false,
30
+ "should_evaluate": false,
31
+ "should_log": false,
32
+ "should_save": true,
33
+ "should_training_stop": false
34
+ },
35
+ "attributes": {}
36
+ }
37
+ },
38
+ "total_flos": 3022884118846464.0,
39
+ "train_batch_size": 32,
40
+ "trial_name": null,
41
+ "trial_params": null
42
+ }
gorkem-pii-ready/TokenBased-BERT/checkpoint-362/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c89573ad34be37ec1bd0425d17a9ec938dcb4bb1682cdabd338ef2e2452dff4
3
+ size 5201
gorkem-pii-ready/TokenBased-BERT/checkpoint-724/config.json ADDED
@@ -0,0 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "BertForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": null,
8
+ "classifier_dropout": null,
9
+ "directionality": "bidi",
10
+ "dtype": "float32",
11
+ "eos_token_id": null,
12
+ "hidden_act": "gelu",
13
+ "hidden_dropout_prob": 0.1,
14
+ "hidden_size": 768,
15
+ "id2label": {
16
+ "0": "O",
17
+ "1": "B-ADDRESS",
18
+ "2": "I-ADDRESS",
19
+ "3": "B-CREDIT_CARD",
20
+ "4": "I-CREDIT_CARD",
21
+ "5": "B-CRYPTO_ADDRESS",
22
+ "6": "I-CRYPTO_ADDRESS",
23
+ "7": "B-EMAIL",
24
+ "8": "I-EMAIL",
25
+ "9": "B-IBAN",
26
+ "10": "I-IBAN",
27
+ "11": "B-OFF_PLATFORM_ATTEMPT",
28
+ "12": "I-OFF_PLATFORM_ATTEMPT",
29
+ "13": "B-PHONE",
30
+ "14": "I-PHONE",
31
+ "15": "B-SOCIAL_MEDIA",
32
+ "16": "I-SOCIAL_MEDIA",
33
+ "17": "B-URL",
34
+ "18": "I-URL"
35
+ },
36
+ "initializer_range": 0.02,
37
+ "intermediate_size": 3072,
38
+ "is_decoder": false,
39
+ "label2id": {
40
+ "B-ADDRESS": 1,
41
+ "B-CREDIT_CARD": 3,
42
+ "B-CRYPTO_ADDRESS": 5,
43
+ "B-EMAIL": 7,
44
+ "B-IBAN": 9,
45
+ "B-OFF_PLATFORM_ATTEMPT": 11,
46
+ "B-PHONE": 13,
47
+ "B-SOCIAL_MEDIA": 15,
48
+ "B-URL": 17,
49
+ "I-ADDRESS": 2,
50
+ "I-CREDIT_CARD": 4,
51
+ "I-CRYPTO_ADDRESS": 6,
52
+ "I-EMAIL": 8,
53
+ "I-IBAN": 10,
54
+ "I-OFF_PLATFORM_ATTEMPT": 12,
55
+ "I-PHONE": 14,
56
+ "I-SOCIAL_MEDIA": 16,
57
+ "I-URL": 18,
58
+ "O": 0
59
+ },
60
+ "layer_norm_eps": 1e-12,
61
+ "max_position_embeddings": 512,
62
+ "model_type": "bert",
63
+ "num_attention_heads": 12,
64
+ "num_hidden_layers": 12,
65
+ "pad_token_id": 0,
66
+ "pooler_fc_size": 768,
67
+ "pooler_num_attention_heads": 12,
68
+ "pooler_num_fc_layers": 3,
69
+ "pooler_size_per_head": 128,
70
+ "pooler_type": "first_token_transform",
71
+ "tie_word_embeddings": true,
72
+ "transformers_version": "5.3.0",
73
+ "type_vocab_size": 2,
74
+ "use_cache": false,
75
+ "vocab_size": 119547
76
+ }
gorkem-pii-ready/TokenBased-BERT/checkpoint-724/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:467894a7fd6a290c0b75cee0d5369f4f1aecb358096fe9c8fa0830eafa7d17d2
3
+ size 709133172
gorkem-pii-ready/TokenBased-BERT/checkpoint-724/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bdf1978b045759595a563309c0bfd53d3c37de65b22ffb07c3552bbd3c9f01f8
3
+ size 1418389387
gorkem-pii-ready/TokenBased-BERT/checkpoint-724/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94474b04ff43a17fb661c45d673a6ef260d5387544fa1700887482935933aae6
3
+ size 14645
gorkem-pii-ready/TokenBased-BERT/checkpoint-724/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9cbc7ec379db25fb0e3c1af9a0587d4b7e17ba3d307149d6600316beb5243ddc
3
+ size 1465
gorkem-pii-ready/TokenBased-BERT/checkpoint-724/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
gorkem-pii-ready/TokenBased-BERT/checkpoint-724/tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "cls_token": "[CLS]",
5
+ "do_lower_case": false,
6
+ "is_local": false,
7
+ "mask_token": "[MASK]",
8
+ "model_max_length": 512,
9
+ "pad_token": "[PAD]",
10
+ "sep_token": "[SEP]",
11
+ "strip_accents": null,
12
+ "tokenize_chinese_chars": true,
13
+ "tokenizer_class": "BertTokenizer",
14
+ "unk_token": "[UNK]"
15
+ }
gorkem-pii-ready/TokenBased-BERT/checkpoint-724/trainer_state.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 724,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "eval_loss": 0.023628847673535347,
15
+ "eval_runtime": 7.6533,
16
+ "eval_samples_per_second": 265.376,
17
+ "eval_steps_per_second": 33.188,
18
+ "step": 362
19
+ },
20
+ {
21
+ "epoch": 1.3812154696132597,
22
+ "grad_norm": 0.11147041618824005,
23
+ "learning_rate": 1.0810313075506448e-05,
24
+ "loss": 0.048348953247070316,
25
+ "step": 500
26
+ },
27
+ {
28
+ "epoch": 2.0,
29
+ "eval_loss": 0.01673239655792713,
30
+ "eval_runtime": 7.3819,
31
+ "eval_samples_per_second": 275.134,
32
+ "eval_steps_per_second": 34.409,
33
+ "step": 724
34
+ }
35
+ ],
36
+ "logging_steps": 500,
37
+ "max_steps": 1086,
38
+ "num_input_tokens_seen": 0,
39
+ "num_train_epochs": 3,
40
+ "save_steps": 500,
41
+ "stateful_callbacks": {
42
+ "TrainerControl": {
43
+ "args": {
44
+ "should_epoch_stop": false,
45
+ "should_evaluate": false,
46
+ "should_log": false,
47
+ "should_save": true,
48
+ "should_training_stop": false
49
+ },
50
+ "attributes": {}
51
+ }
52
+ },
53
+ "total_flos": 6045768237692928.0,
54
+ "train_batch_size": 32,
55
+ "trial_name": null,
56
+ "trial_params": null
57
+ }
gorkem-pii-ready/TokenBased-BERT/checkpoint-724/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c89573ad34be37ec1bd0425d17a9ec938dcb4bb1682cdabd338ef2e2452dff4
3
+ size 5201
gorkem-pii-ready/TokenBased-BERT/config.json ADDED
@@ -0,0 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "BertForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": null,
8
+ "classifier_dropout": null,
9
+ "directionality": "bidi",
10
+ "dtype": "float32",
11
+ "eos_token_id": null,
12
+ "hidden_act": "gelu",
13
+ "hidden_dropout_prob": 0.1,
14
+ "hidden_size": 768,
15
+ "id2label": {
16
+ "0": "O",
17
+ "1": "B-ADDRESS",
18
+ "2": "I-ADDRESS",
19
+ "3": "B-CREDIT_CARD",
20
+ "4": "I-CREDIT_CARD",
21
+ "5": "B-CRYPTO_ADDRESS",
22
+ "6": "I-CRYPTO_ADDRESS",
23
+ "7": "B-EMAIL",
24
+ "8": "I-EMAIL",
25
+ "9": "B-IBAN",
26
+ "10": "I-IBAN",
27
+ "11": "B-OFF_PLATFORM_ATTEMPT",
28
+ "12": "I-OFF_PLATFORM_ATTEMPT",
29
+ "13": "B-PHONE",
30
+ "14": "I-PHONE",
31
+ "15": "B-SOCIAL_MEDIA",
32
+ "16": "I-SOCIAL_MEDIA",
33
+ "17": "B-URL",
34
+ "18": "I-URL"
35
+ },
36
+ "initializer_range": 0.02,
37
+ "intermediate_size": 3072,
38
+ "is_decoder": false,
39
+ "label2id": {
40
+ "B-ADDRESS": 1,
41
+ "B-CREDIT_CARD": 3,
42
+ "B-CRYPTO_ADDRESS": 5,
43
+ "B-EMAIL": 7,
44
+ "B-IBAN": 9,
45
+ "B-OFF_PLATFORM_ATTEMPT": 11,
46
+ "B-PHONE": 13,
47
+ "B-SOCIAL_MEDIA": 15,
48
+ "B-URL": 17,
49
+ "I-ADDRESS": 2,
50
+ "I-CREDIT_CARD": 4,
51
+ "I-CRYPTO_ADDRESS": 6,
52
+ "I-EMAIL": 8,
53
+ "I-IBAN": 10,
54
+ "I-OFF_PLATFORM_ATTEMPT": 12,
55
+ "I-PHONE": 14,
56
+ "I-SOCIAL_MEDIA": 16,
57
+ "I-URL": 18,
58
+ "O": 0
59
+ },
60
+ "layer_norm_eps": 1e-12,
61
+ "max_position_embeddings": 512,
62
+ "model_type": "bert",
63
+ "num_attention_heads": 12,
64
+ "num_hidden_layers": 12,
65
+ "pad_token_id": 0,
66
+ "pooler_fc_size": 768,
67
+ "pooler_num_attention_heads": 12,
68
+ "pooler_num_fc_layers": 3,
69
+ "pooler_size_per_head": 128,
70
+ "pooler_type": "first_token_transform",
71
+ "tie_word_embeddings": true,
72
+ "transformers_version": "5.3.0",
73
+ "type_vocab_size": 2,
74
+ "use_cache": false,
75
+ "vocab_size": 119547
76
+ }
gorkem-pii-ready/TokenBased-BERT/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97f5839fe9e4c978611fb55c0a0c3bc06b9faebb1c9e37a05c86bf7ed2ef729b
3
+ size 709133172
gorkem-pii-ready/TokenBased-BERT/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
gorkem-pii-ready/TokenBased-BERT/tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "cls_token": "[CLS]",
5
+ "do_lower_case": false,
6
+ "is_local": false,
7
+ "mask_token": "[MASK]",
8
+ "model_max_length": 512,
9
+ "pad_token": "[PAD]",
10
+ "sep_token": "[SEP]",
11
+ "strip_accents": null,
12
+ "tokenize_chinese_chars": true,
13
+ "tokenizer_class": "BertTokenizer",
14
+ "unk_token": "[UNK]"
15
+ }
gorkem-pii-ready/TokenBased-BERT/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c89573ad34be37ec1bd0425d17a9ec938dcb4bb1682cdabd338ef2e2452dff4
3
+ size 5201