Hugging Face's logo

PuxAI
/
GenPII-framework-Data

Safetensors
Model card Files
xet
 Community
PuxAI commited on
Commit
7239a3f
·
verified ·
1 Parent(s): fcb4102

Upload folder using huggingface_hub

Browse files
Files changed (11) hide show
  1. financial-pii-ready/TokenBased-CRF/checkpoint-63/config.json +66 -22
  2. financial-pii-ready/TokenBased-CRF/checkpoint-63/model.safetensors +2 -2
  3. financial-pii-ready/TokenBased-CRF/checkpoint-63/optimizer.pt +2 -2
  4. financial-pii-ready/TokenBased-CRF/checkpoint-63/rng_state.pth +1 -1
  5. financial-pii-ready/TokenBased-CRF/checkpoint-63/tokenizer.json +0 -0
  6. financial-pii-ready/TokenBased-CRF/checkpoint-63/tokenizer_config.json +7 -5
  7. financial-pii-ready/TokenBased-CRF/checkpoint-63/trainer_state.json +4 -4
  8. financial-pii-ready/TokenBased-CRF/config.json +66 -22
  9. financial-pii-ready/TokenBased-CRF/pytorch_model.bin +2 -2
  10. financial-pii-ready/TokenBased-CRF/tokenizer.json +0 -0
  11. financial-pii-ready/TokenBased-CRF/tokenizer_config.json +7 -5
financial-pii-ready/TokenBased-CRF/checkpoint-63/config.json CHANGED
@@ -1,40 +1,84 @@
1
  {
2
- "add_cross_attention": false,
3
  "architectures": [
4
  "TransformerCrfForTokenClassification"
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
  },
 
18
  "initializer_range": 0.02,
19
- "intermediate_size": 3072,
20
- "is_decoder": false,
21
  "label2id": {
22
  "O": 0
23
  },
24
- "layer_norm_eps": 1e-12,
25
- "max_position_embeddings": 512,
26
- "model_type": "bert",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
27
  "num_attention_heads": 12,
28
- "num_hidden_layers": 12,
29
- "pad_token_id": 0,
30
- "pooler_fc_size": 768,
31
- "pooler_num_attention_heads": 12,
32
- "pooler_num_fc_layers": 3,
33
- "pooler_size_per_head": 128,
34
- "pooler_type": "first_token_transform",
 
 
 
 
 
 
 
 
 
35
  "tie_word_embeddings": true,
36
  "transformers_version": "5.3.0",
37
- "type_vocab_size": 2,
38
  "use_cache": false,
39
- "vocab_size": 119547
40
  }
 
1
  {
 
2
  "architectures": [
3
  "TransformerCrfForTokenClassification"
4
  ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 50281,
8
+ "classifier_activation": "gelu",
9
+ "classifier_bias": false,
10
+ "classifier_dropout": 0.0,
11
+ "classifier_pooling": "mean",
12
+ "cls_token_id": 50281,
13
+ "decoder_bias": true,
14
+ "deterministic_flash_attn": false,
15
  "dtype": "float32",
16
+ "embedding_dropout": 0.0,
17
+ "eos_token_id": 50282,
18
+ "global_attn_every_n_layers": 3,
19
+ "gradient_checkpointing": false,
20
+ "hidden_activation": "gelu",
21
  "hidden_size": 768,
22
  "id2label": {
23
  "0": "O"
24
  },
25
+ "initializer_cutoff_factor": 2.0,
26
  "initializer_range": 0.02,
27
+ "intermediate_size": 1152,
 
28
  "label2id": {
29
  "O": 0
30
  },
31
+ "layer_norm_eps": 1e-05,
32
+ "layer_types": [
33
+ "full_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "full_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "full_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "full_attention",
49
+ "sliding_attention",
50
+ "sliding_attention",
51
+ "full_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "full_attention"
55
+ ],
56
+ "local_attention": 128,
57
+ "max_position_embeddings": 8192,
58
+ "mlp_bias": false,
59
+ "mlp_dropout": 0.0,
60
+ "model_type": "modernbert",
61
+ "norm_bias": false,
62
+ "norm_eps": 1e-05,
63
  "num_attention_heads": 12,
64
+ "num_hidden_layers": 22,
65
+ "pad_token_id": 50283,
66
+ "position_embedding_type": "absolute",
67
+ "rope_parameters": {
68
+ "full_attention": {
69
+ "rope_theta": 160000.0,
70
+ "rope_type": "default"
71
+ },
72
+ "sliding_attention": {
73
+ "rope_theta": 10000.0,
74
+ "rope_type": "default"
75
+ }
76
+ },
77
+ "sep_token_id": 50282,
78
+ "sparse_pred_ignore_index": -100,
79
+ "sparse_prediction": false,
80
  "tie_word_embeddings": true,
81
  "transformers_version": "5.3.0",
 
82
  "use_cache": false,
83
+ "vocab_size": 50368
84
  }
financial-pii-ready/TokenBased-CRF/checkpoint-63/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1a33fa8f20989e1678a3edc77b7b2fddfda14c9f0fb76b514af560bf54babbac
3
- size 711441992
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ad7634dceddbed7c5bfe64f2538d7f71bfaef155afaa3081e82e8902568a8cb
3
+ size 596075208
financial-pii-ready/TokenBased-CRF/checkpoint-63/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:79148412b64f6a2290a7249211de11a1587d63b7a027284d03c3ba7d2f16c7e8
3
- size 1418281163
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9bf6b24ce637aeb60e01e75cab68f8b57f7873fc864167c98da41fbdcf9525d9
3
+ size 1192239755
financial-pii-ready/TokenBased-CRF/checkpoint-63/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f39eab8ed980549bfffcd8b948e8852bef979d820a1c47a898b2c9f270cc3986
3
  size 14645
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08504cb4efee454165d24aa00ad8e58713d98093f38b30f8a5112618d186bfa1
3
  size 14645
financial-pii-ready/TokenBased-CRF/checkpoint-63/tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
financial-pii-ready/TokenBased-CRF/checkpoint-63/tokenizer_config.json CHANGED
@@ -1,15 +1,17 @@
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
  }
 
1
  {
2
  "add_prefix_space": true,
3
  "backend": "tokenizers",
4
+ "clean_up_tokenization_spaces": true,
5
  "cls_token": "[CLS]",
 
6
  "is_local": false,
7
  "mask_token": "[MASK]",
8
+ "model_input_names": [
9
+ "input_ids",
10
+ "attention_mask"
11
+ ],
12
+ "model_max_length": 8192,
13
  "pad_token": "[PAD]",
14
  "sep_token": "[SEP]",
15
+ "tokenizer_class": "TokenizersBackend",
 
 
16
  "unk_token": "[UNK]"
17
  }
financial-pii-ready/TokenBased-CRF/checkpoint-63/trainer_state.json CHANGED
@@ -12,9 +12,9 @@
12
  {
13
  "epoch": 1.0,
14
  "eval_loss": 0.0,
15
- "eval_runtime": 0.5944,
16
- "eval_samples_per_second": 50.475,
17
- "eval_steps_per_second": 6.73,
18
  "step": 63
19
  }
20
  ],
@@ -35,7 +35,7 @@
35
  "attributes": {}
36
  }
37
  },
38
- "total_flos": 131554351104000.0,
39
  "train_batch_size": 8,
40
  "trial_name": null,
41
  "trial_params": null
 
12
  {
13
  "epoch": 1.0,
14
  "eval_loss": 0.0,
15
+ "eval_runtime": 0.6318,
16
+ "eval_samples_per_second": 47.484,
17
+ "eval_steps_per_second": 6.331,
18
  "step": 63
19
  }
20
  ],
 
35
  "attributes": {}
36
  }
37
  },
38
+ "total_flos": 169470597120000.0,
39
  "train_batch_size": 8,
40
  "trial_name": null,
41
  "trial_params": null
financial-pii-ready/TokenBased-CRF/config.json CHANGED
@@ -1,40 +1,84 @@
1
  {
2
- "add_cross_attention": false,
3
  "architectures": [
4
  "TransformerCrfForTokenClassification"
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
  },
 
18
  "initializer_range": 0.02,
19
- "intermediate_size": 3072,
20
- "is_decoder": false,
21
  "label2id": {
22
  "O": 0
23
  },
24
- "layer_norm_eps": 1e-12,
25
- "max_position_embeddings": 512,
26
- "model_type": "bert",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
27
  "num_attention_heads": 12,
28
- "num_hidden_layers": 12,
29
- "pad_token_id": 0,
30
- "pooler_fc_size": 768,
31
- "pooler_num_attention_heads": 12,
32
- "pooler_num_fc_layers": 3,
33
- "pooler_size_per_head": 128,
34
- "pooler_type": "first_token_transform",
 
 
 
 
 
 
 
 
 
35
  "tie_word_embeddings": true,
36
  "transformers_version": "5.3.0",
37
- "type_vocab_size": 2,
38
  "use_cache": false,
39
- "vocab_size": 119547
40
  }
 
1
  {
 
2
  "architectures": [
3
  "TransformerCrfForTokenClassification"
4
  ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 50281,
8
+ "classifier_activation": "gelu",
9
+ "classifier_bias": false,
10
+ "classifier_dropout": 0.0,
11
+ "classifier_pooling": "mean",
12
+ "cls_token_id": 50281,
13
+ "decoder_bias": true,
14
+ "deterministic_flash_attn": false,
15
  "dtype": "float32",
16
+ "embedding_dropout": 0.0,
17
+ "eos_token_id": 50282,
18
+ "global_attn_every_n_layers": 3,
19
+ "gradient_checkpointing": false,
20
+ "hidden_activation": "gelu",
21
  "hidden_size": 768,
22
  "id2label": {
23
  "0": "O"
24
  },
25
+ "initializer_cutoff_factor": 2.0,
26
  "initializer_range": 0.02,
27
+ "intermediate_size": 1152,
 
28
  "label2id": {
29
  "O": 0
30
  },
31
+ "layer_norm_eps": 1e-05,
32
+ "layer_types": [
33
+ "full_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "full_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "full_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "full_attention",
49
+ "sliding_attention",
50
+ "sliding_attention",
51
+ "full_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "full_attention"
55
+ ],
56
+ "local_attention": 128,
57
+ "max_position_embeddings": 8192,
58
+ "mlp_bias": false,
59
+ "mlp_dropout": 0.0,
60
+ "model_type": "modernbert",
61
+ "norm_bias": false,
62
+ "norm_eps": 1e-05,
63
  "num_attention_heads": 12,
64
+ "num_hidden_layers": 22,
65
+ "pad_token_id": 50283,
66
+ "position_embedding_type": "absolute",
67
+ "rope_parameters": {
68
+ "full_attention": {
69
+ "rope_theta": 160000.0,
70
+ "rope_type": "default"
71
+ },
72
+ "sliding_attention": {
73
+ "rope_theta": 10000.0,
74
+ "rope_type": "default"
75
+ }
76
+ },
77
+ "sep_token_id": 50282,
78
+ "sparse_pred_ignore_index": -100,
79
+ "sparse_prediction": false,
80
  "tie_word_embeddings": true,
81
  "transformers_version": "5.3.0",
 
82
  "use_cache": false,
83
+ "vocab_size": 50368
84
  }
financial-pii-ready/TokenBased-CRF/pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f62b92b941f41460837da75a5cc8e4a3ca9b95fa49700be682eb87c0ff17c349
3
- size 711504083
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed31be5bbaf8407607ebea72a793e7d2042ffa9a758ccc645cd1b60b48e9b74b
3
+ size 596121359
financial-pii-ready/TokenBased-CRF/tokenizer.json CHANGED
The diff for this file is too large to render. See raw diff
 
financial-pii-ready/TokenBased-CRF/tokenizer_config.json CHANGED
@@ -1,15 +1,17 @@
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
  }
 
1
  {
2
  "add_prefix_space": true,
3
  "backend": "tokenizers",
4
+ "clean_up_tokenization_spaces": true,
5
  "cls_token": "[CLS]",
 
6
  "is_local": false,
7
  "mask_token": "[MASK]",
8
+ "model_input_names": [
9
+ "input_ids",
10
+ "attention_mask"
11
+ ],
12
+ "model_max_length": 8192,
13
  "pad_token": "[PAD]",
14
  "sep_token": "[SEP]",
15
+ "tokenizer_class": "TokenizersBackend",
 
 
16
  "unk_token": "[UNK]"
17
  }