Hugging Face's logo

capazme
/
privacy-filter-it

Token Classification
Safetensors
Italian
opf
privacy_filter
privacy-filter
pii-detection
italian
anonymization
ner
Model card Files
xet
 Community
capazme commited on
Commit
041b204
·
verified ·
1 Parent(s): 18c887e

Upload fine-tuned checkpoint: checkpoint_step1_italian_docs_v2

Browse files
Files changed (5) hide show
  1. README.md +127 -3
  2. USAGE.txt +7 -0
  3. config.json +124 -0
  4. finetune_summary.json +261 -0
  5. model.safetensors +3 -0
README.md CHANGED
@@ -1,3 +1,127 @@
1
- ---
2
- license: gpl
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - it
4
+ license: apache-2.0
5
+ library_name: opf
6
+ base_model: openai/privacy-filter
7
+ pipeline_tag: token-classification
8
+ tags:
9
+ - privacy-filter
10
+ - pii-detection
11
+ - italian
12
+ - anonymization
13
+ - ner
14
+ - opf
15
+ ---
16
+
17
+ # privacy-filter-it
18
+
19
+ Fine-tuning di [openai/privacy-filter](https://huggingface.co/openai/privacy-filter) su documenti italiani sintetici per il riconoscimento di PII (Personally Identifiable Information).
20
+
21
+ Modello addestrato su dataset sintetico italiano (checkpoint_step1_italian_docs_v2).
22
+
23
+ ## ⚠️ Come caricare il modello
24
+
25
+ Questo modello usa un'architettura **custom** (`model_type: privacy_filter`) **non** registrata in `transformers`. NON funziona con `AutoModel` / `transformers.pipeline`.
26
+
27
+ Per usarlo serve la libreria `opf`:
28
+
29
+ ```bash
30
+ pip install git+https://github.com/openai/privacy-filter.git
31
+ ```
32
+
33
+ ```python
34
+ import os
35
+ os.environ['OPF_MOE_TRITON'] = '0' # disabilita kernel CUDA-only su MPS/CPU
36
+
37
+ from opf import OPF
38
+ from huggingface_hub import snapshot_download
39
+
40
+ # Scarica il modello (viene messo in cache locale)
41
+ local_path = snapshot_download(repo_id='capazme/privacy-filter-it')
42
+
43
+ model = OPF(
44
+ model=local_path,
45
+ device='cuda', # oppure 'mps' (Apple Silicon) o 'cpu'
46
+ output_mode='typed',
47
+ decode_mode='viterbi',
48
+ )
49
+
50
+ text = 'Il sottoscritto Mario Rossi, CF RSSMRA80A01H501U, residente in Via Roma 10, Milano.'
51
+ result = model.redact(text)
52
+
53
+ print(result.redacted_text)
54
+ # -> Il sottoscritto <PRIVATE_PERSON>, <CODICE_FISCALE>, residente in <PRIVATE_ADDRESS>.
55
+
56
+ for span in result.detected_spans:
57
+ print(f'{span.label:25s} "{span.text}" [{span.start}:{span.end}]')
58
+ ```
59
+
60
+ ## 📋 Categorie riconosciute
61
+
62
+ Il modello riconosce **18** categorie di PII italiane:
63
+
64
+ | Categoria | Descrizione |
65
+ |---|---|
66
+ | `private_person` | Nomi di persone fisiche |
67
+ | `private_address` | Indirizzi (vie, città, numeri civici) |
68
+ | `private_email` | Indirizzi email |
69
+ | `private_phone` | Numeri di telefono italiani |
70
+ | `private_url` | URL contenenti dati personali |
71
+ | `private_date` | Date (nascita, scadenze, eventi) |
72
+ | `account_number` | Numeri di conto (generici) |
73
+ | `secret` | Credenziali, password, token |
74
+ | `codice_fiscale` | Codice Fiscale italiano (16 caratteri) |
75
+ | `carta_identita` | Numero Carta d'Identità italiana |
76
+ | `patente` | Numero Patente di guida |
77
+ | `passaporto` | Numero Passaporto |
78
+ | `partita_iva` | Partita IVA italiana (11 cifre) |
79
+ | `iban` | IBAN italiano (27 caratteri) |
80
+ | `tessera_sanitaria` | Tessera Sanitaria |
81
+ | `numero_procedimento` | Numero procedimento legale (RG) |
82
+ | `riferimento_catastale` | Riferimento catastale (foglio/mappale) |
83
+ | `parte_in_causa` | Parti in procedimento giudiziario |
84
+
85
+ ## 📊 Dettagli training
86
+
87
+ - **Base model**: `openai/privacy-filter`
88
+ - **Dataset**: sintetico, generato dal modulo `dataset_builder.py` (vedi [repo GitHub](https://github.com/capazme/privacy-filter-it))
89
+ - **Dati**: **7500** esempi di training, **1250** di validation, **(held-out, non usato in training)** di test (held-out)
90
+ - **Epoche**: 14
91
+ - **Batch size**: 1
92
+ - **Grad accum steps**: 4
93
+ - **Learning rate**: 1e-05
94
+ - **Hardware training**: cuda
95
+ - **Best epoch**: 14 (validation loss: 0.0000)
96
+ - **Param dtype**: bfloat16
97
+
98
+ ## 🎯 Metriche (validation set)
99
+
100
+ - **Token accuracy** (best): 1.0000
101
+ - **Validation loss** (best): 0.0000
102
+
103
+ ## 🎨 Esempi di output
104
+
105
+ **Input**: `Per bonifici IBAN IT60X0542811101000000123456 intestato a Luigi Bianchi. luigi.bianchi@studio.it`
106
+
107
+ **Output**: `Per bonifici <IBAN> intestato a <PRIVATE_PERSON>. <PRIVATE_EMAIL>`
108
+
109
+ ## ⚖️ Licenza & limitazioni
110
+
111
+ - **Licenza**: Apache 2.0 (ereditata dal base model)
112
+ - **Limiti**: il dataset è sintetico — il modello potrebbe avere pattern overfitted su formati tipici (es. "CF RSSMRA80A01H501U" preceduto da prefisso). Testa con i tuoi testi prima dell'uso in produzione.
113
+ - **Contesto**: addestrato su testo italiano generico (email, CV, news, chat, business). Non ottimizzato per domini specifici (medico, scientifico, etc.).
114
+ - **Dati sintetici**: nessun dato reale di terze parti usato nel training. Tutti gli esempi sono generati programmaticamente con formati italiani validi ma valori casuali.
115
+
116
+ ## 📎 Citazione
117
+
118
+ Se usi questo modello, per favore cita il lavoro originale di OpenAI:
119
+
120
+ ```
121
+ @misc{openai-privacy-filter,
122
+ title = {Privacy Filter},
123
+ author = {OpenAI},
124
+ year = {2024},
125
+ url = {https://github.com/openai/privacy-filter}
126
+ }
127
+ ```
USAGE.txt ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ Finetuned checkpoint generated by `opf train`.
2
+
3
+ Run local inference:
4
+ opf --checkpoint /kaggle/working/checkpoint_step1_italian_docs_v2 --device cuda "Alice was born on 1990-01-02."
5
+
6
+ Run eval:
7
+ opf eval /path/to/eval.jsonl --checkpoint /kaggle/working/checkpoint_step1_italian_docs_v2 --device cuda
config.json ADDED
@@ -0,0 +1,124 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bidirectional_context": true,
3
+ "bidirectional_left_context": 128,
4
+ "bidirectional_right_context": 128,
5
+ "category_version": "italian_legal_v1",
6
+ "default_n_ctx": 128000,
7
+ "encoding": "o200k_base",
8
+ "experts_per_token": 4,
9
+ "head_dim": 64,
10
+ "hidden_size": 640,
11
+ "inference_contract_version": 1,
12
+ "initial_context_length": 4096,
13
+ "intermediate_size": 640,
14
+ "max_position_embeddings": 131072,
15
+ "model_type": "privacy_filter",
16
+ "ner_class_names": [
17
+ "O",
18
+ "B-private_person",
19
+ "I-private_person",
20
+ "E-private_person",
21
+ "S-private_person",
22
+ "B-private_address",
23
+ "I-private_address",
24
+ "E-private_address",
25
+ "S-private_address",
26
+ "B-private_email",
27
+ "I-private_email",
28
+ "E-private_email",
29
+ "S-private_email",
30
+ "B-private_phone",
31
+ "I-private_phone",
32
+ "E-private_phone",
33
+ "S-private_phone",
34
+ "B-private_url",
35
+ "I-private_url",
36
+ "E-private_url",
37
+ "S-private_url",
38
+ "B-private_date",
39
+ "I-private_date",
40
+ "E-private_date",
41
+ "S-private_date",
42
+ "B-account_number",
43
+ "I-account_number",
44
+ "E-account_number",
45
+ "S-account_number",
46
+ "B-secret",
47
+ "I-secret",
48
+ "E-secret",
49
+ "S-secret",
50
+ "B-codice_fiscale",
51
+ "I-codice_fiscale",
52
+ "E-codice_fiscale",
53
+ "S-codice_fiscale",
54
+ "B-carta_identita",
55
+ "I-carta_identita",
56
+ "E-carta_identita",
57
+ "S-carta_identita",
58
+ "B-patente",
59
+ "I-patente",
60
+ "E-patente",
61
+ "S-patente",
62
+ "B-passaporto",
63
+ "I-passaporto",
64
+ "E-passaporto",
65
+ "S-passaporto",
66
+ "B-partita_iva",
67
+ "I-partita_iva",
68
+ "E-partita_iva",
69
+ "S-partita_iva",
70
+ "B-iban",
71
+ "I-iban",
72
+ "E-iban",
73
+ "S-iban",
74
+ "B-tessera_sanitaria",
75
+ "I-tessera_sanitaria",
76
+ "E-tessera_sanitaria",
77
+ "S-tessera_sanitaria",
78
+ "B-numero_procedimento",
79
+ "I-numero_procedimento",
80
+ "E-numero_procedimento",
81
+ "S-numero_procedimento",
82
+ "B-riferimento_catastale",
83
+ "I-riferimento_catastale",
84
+ "E-riferimento_catastale",
85
+ "S-riferimento_catastale",
86
+ "B-parte_in_causa",
87
+ "I-parte_in_causa",
88
+ "E-parte_in_causa",
89
+ "S-parte_in_causa"
90
+ ],
91
+ "num_attention_heads": 14,
92
+ "num_experts": 128,
93
+ "num_hidden_layers": 8,
94
+ "num_key_value_heads": 2,
95
+ "num_labels": 73,
96
+ "param_dtype": "bfloat16",
97
+ "rope_ntk_alpha": 1.0,
98
+ "rope_ntk_beta": 32.0,
99
+ "rope_scaling_factor": 32.0,
100
+ "rope_theta": 150000,
101
+ "sliding_window": 257,
102
+ "span_class_names": [
103
+ "O",
104
+ "private_person",
105
+ "private_address",
106
+ "private_email",
107
+ "private_phone",
108
+ "private_url",
109
+ "private_date",
110
+ "account_number",
111
+ "secret",
112
+ "codice_fiscale",
113
+ "carta_identita",
114
+ "patente",
115
+ "passaporto",
116
+ "partita_iva",
117
+ "iban",
118
+ "tessera_sanitaria",
119
+ "numero_procedimento",
120
+ "riferimento_catastale",
121
+ "parte_in_causa"
122
+ ],
123
+ "vocab_size": 200064
124
+ }
finetune_summary.json ADDED
@@ -0,0 +1,261 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "artifacts": {
3
+ "config_json": "/kaggle/working/checkpoint_step1_italian_docs_v2/config.json",
4
+ "model_safetensors": "/kaggle/working/checkpoint_step1_italian_docs_v2/model.safetensors",
5
+ "summary_json": "/kaggle/working/checkpoint_step1_italian_docs_v2/finetune_summary.json"
6
+ },
7
+ "base_checkpoint": "/root/.opf/privacy_filter",
8
+ "batch_size": 1,
9
+ "best_epoch": 14,
10
+ "best_metric": 1.256955582021411e-06,
11
+ "best_metric_name": "validation_loss",
12
+ "checkpoint_category_version": "v2",
13
+ "device": "cuda",
14
+ "elapsed_s": 8991.771278478,
15
+ "encoding": "o200k_base",
16
+ "epoch_metrics": [
17
+ {
18
+ "elapsed_s": 611.6480500910002,
19
+ "epoch": 1,
20
+ "optimizer_steps": 1875,
21
+ "train_batches": 7500,
22
+ "train_loss": 0.18406920914817018,
23
+ "train_token_accuracy": 0.9626421135608234,
24
+ "train_tokens": 246106,
25
+ "validation_batches": 1250,
26
+ "validation_loss": 0.0010319853343994626,
27
+ "validation_token_accuracy": 0.9997995087965516,
28
+ "validation_tokens": 39902
29
+ },
30
+ {
31
+ "elapsed_s": 603.1925267070001,
32
+ "epoch": 2,
33
+ "optimizer_steps": 1875,
34
+ "train_batches": 7500,
35
+ "train_loss": 0.0006751688251858842,
36
+ "train_token_accuracy": 0.9999309240733668,
37
+ "train_tokens": 246106,
38
+ "validation_batches": 1250,
39
+ "validation_loss": 0.0004481305598764045,
40
+ "validation_token_accuracy": 0.9998746929978447,
41
+ "validation_tokens": 39902
42
+ },
43
+ {
44
+ "elapsed_s": 595.9333531040002,
45
+ "epoch": 3,
46
+ "optimizer_steps": 1875,
47
+ "train_batches": 7500,
48
+ "train_loss": 0.0004284777879854081,
49
+ "train_token_accuracy": 0.9999349873631688,
50
+ "train_tokens": 246106,
51
+ "validation_batches": 1250,
52
+ "validation_loss": 0.0011235129029634563,
53
+ "validation_token_accuracy": 0.9997995087965516,
54
+ "validation_tokens": 39902
55
+ },
56
+ {
57
+ "elapsed_s": 594.8147814399999,
58
+ "epoch": 4,
59
+ "optimizer_steps": 1875,
60
+ "train_batches": 7500,
61
+ "train_loss": 0.0004589788315978451,
62
+ "train_token_accuracy": 0.999910607624357,
63
+ "train_tokens": 246106,
64
+ "validation_batches": 1250,
65
+ "validation_loss": 0.0005426766982118603,
66
+ "validation_token_accuracy": 0.9997995087965516,
67
+ "validation_tokens": 39902
68
+ },
69
+ {
70
+ "elapsed_s": 594.5084605920001,
71
+ "epoch": 5,
72
+ "optimizer_steps": 1875,
73
+ "train_batches": 7500,
74
+ "train_loss": 0.00025950950850595604,
75
+ "train_token_accuracy": 0.9999431139427726,
76
+ "train_tokens": 246106,
77
+ "validation_batches": 1250,
78
+ "validation_loss": 0.0006384832375886648,
79
+ "validation_token_accuracy": 0.9998997543982758,
80
+ "validation_tokens": 39902
81
+ },
82
+ {
83
+ "elapsed_s": 601.9116796880007,
84
+ "epoch": 6,
85
+ "optimizer_steps": 1875,
86
+ "train_batches": 7500,
87
+ "train_loss": 0.00033364914724159483,
88
+ "train_token_accuracy": 0.9999390506529706,
89
+ "train_tokens": 246106,
90
+ "validation_batches": 1250,
91
+ "validation_loss": 0.00023689508041231628,
92
+ "validation_token_accuracy": 0.9999498771991379,
93
+ "validation_tokens": 39902
94
+ },
95
+ {
96
+ "elapsed_s": 599.1612815619992,
97
+ "epoch": 7,
98
+ "optimizer_steps": 1875,
99
+ "train_batches": 7500,
100
+ "train_loss": 0.00021150960081863236,
101
+ "train_token_accuracy": 0.9999431139427726,
102
+ "train_tokens": 246106,
103
+ "validation_batches": 1250,
104
+ "validation_loss": 0.0001638231382475539,
105
+ "validation_token_accuracy": 0.9999248157987068,
106
+ "validation_tokens": 39902
107
+ },
108
+ {
109
+ "elapsed_s": 595.1891281780008,
110
+ "epoch": 8,
111
+ "optimizer_steps": 1875,
112
+ "train_batches": 7500,
113
+ "train_loss": 0.00011636061623750538,
114
+ "train_token_accuracy": 0.9999837468407922,
115
+ "train_tokens": 246106,
116
+ "validation_batches": 1250,
117
+ "validation_loss": 0.00019882116281728446,
118
+ "validation_token_accuracy": 0.9999248157987068,
119
+ "validation_tokens": 39902
120
+ },
121
+ {
122
+ "elapsed_s": 593.3688169760007,
123
+ "epoch": 9,
124
+ "optimizer_steps": 1875,
125
+ "train_batches": 7500,
126
+ "train_loss": 0.0001755927075592558,
127
+ "train_token_accuracy": 0.9999634303917824,
128
+ "train_tokens": 246106,
129
+ "validation_batches": 1250,
130
+ "validation_loss": 0.00022523166237503278,
131
+ "validation_token_accuracy": 0.9998997543982758,
132
+ "validation_tokens": 39902
133
+ },
134
+ {
135
+ "elapsed_s": 593.3411462200002,
136
+ "epoch": 10,
137
+ "optimizer_steps": 1875,
138
+ "train_batches": 7500,
139
+ "train_loss": 0.0001144706639999654,
140
+ "train_token_accuracy": 0.9999796835509902,
141
+ "train_tokens": 246106,
142
+ "validation_batches": 1250,
143
+ "validation_loss": 0.00022547725828748192,
144
+ "validation_token_accuracy": 0.9999248157987068,
145
+ "validation_tokens": 39902
146
+ },
147
+ {
148
+ "elapsed_s": 603.6394337089987,
149
+ "epoch": 11,
150
+ "optimizer_steps": 1875,
151
+ "train_batches": 7500,
152
+ "train_loss": 0.00017106058540362378,
153
+ "train_token_accuracy": 0.9999634303917824,
154
+ "train_tokens": 246106,
155
+ "validation_batches": 1250,
156
+ "validation_loss": 5.802690870656621e-05,
157
+ "validation_token_accuracy": 1.0,
158
+ "validation_tokens": 39902
159
+ },
160
+ {
161
+ "elapsed_s": 596.6633162110011,
162
+ "epoch": 12,
163
+ "optimizer_steps": 1875,
164
+ "train_batches": 7500,
165
+ "train_loss": 5.1230087411499746e-05,
166
+ "train_token_accuracy": 0.999995936710198,
167
+ "train_tokens": 246106,
168
+ "validation_batches": 1250,
169
+ "validation_loss": 0.0001793956412968626,
170
+ "validation_token_accuracy": 0.999974938599569,
171
+ "validation_tokens": 39902
172
+ },
173
+ {
174
+ "elapsed_s": 597.4834135400015,
175
+ "epoch": 13,
176
+ "optimizer_steps": 1875,
177
+ "train_batches": 7500,
178
+ "train_loss": 9.818313904666025e-05,
179
+ "train_token_accuracy": 0.9999756202611882,
180
+ "train_tokens": 246106,
181
+ "validation_batches": 1250,
182
+ "validation_loss": 0.0001664656747663711,
183
+ "validation_token_accuracy": 0.999974938599569,
184
+ "validation_tokens": 39902
185
+ },
186
+ {
187
+ "elapsed_s": 605.1379507809997,
188
+ "epoch": 14,
189
+ "optimizer_steps": 1875,
190
+ "train_batches": 7500,
191
+ "train_loss": 0.00013058107322193007,
192
+ "train_token_accuracy": 0.9999756202611882,
193
+ "train_tokens": 246106,
194
+ "validation_batches": 1250,
195
+ "validation_loss": 1.256955582021411e-06,
196
+ "validation_token_accuracy": 1.0,
197
+ "validation_tokens": 39902
198
+ },
199
+ {
200
+ "elapsed_s": 597.8637465249994,
201
+ "epoch": 15,
202
+ "optimizer_steps": 1875,
203
+ "train_batches": 7500,
204
+ "train_loss": 3.845093296363575e-06,
205
+ "train_token_accuracy": 1.0,
206
+ "train_tokens": 246106,
207
+ "validation_batches": 1250,
208
+ "validation_loss": 1.754828393192001e-05,
209
+ "validation_token_accuracy": 1.0,
210
+ "validation_tokens": 39902
211
+ }
212
+ ],
213
+ "epochs": 15,
214
+ "generated_at_unix": 1777048760.4942422,
215
+ "grad_accum_steps": 4,
216
+ "label_space_json_path": "/kaggle/working/custom_label_space.json",
217
+ "label_space_source": "label-space-json",
218
+ "learning_rate": 1e-05,
219
+ "max_grad_norm": 1.0,
220
+ "num_output_labels": 73,
221
+ "num_train_examples": 7500,
222
+ "num_train_windows": 7500,
223
+ "num_validation_examples": 1250,
224
+ "num_validation_windows": 1250,
225
+ "output_checkpoint_dir": "/kaggle/working/checkpoint_step1_italian_docs_v2",
226
+ "output_head_reinitialized": true,
227
+ "output_head_rows_copied": 73,
228
+ "output_head_rows_copied_exact": 33,
229
+ "output_head_rows_copied_fallback": 40,
230
+ "resolved_category_version": "italian_legal_v1",
231
+ "resolved_n_ctx": 128000,
232
+ "schema_version": 1,
233
+ "serialized_param_dtype": "bfloat16",
234
+ "span_class_names": [
235
+ "O",
236
+ "private_person",
237
+ "private_address",
238
+ "private_email",
239
+ "private_phone",
240
+ "private_url",
241
+ "private_date",
242
+ "account_number",
243
+ "secret",
244
+ "codice_fiscale",
245
+ "carta_identita",
246
+ "patente",
247
+ "passaporto",
248
+ "partita_iva",
249
+ "iban",
250
+ "tessera_sanitaria",
251
+ "numero_procedimento",
252
+ "riferimento_catastale",
253
+ "parte_in_causa"
254
+ ],
255
+ "train_dataset": "/kaggle/working/datasets/step1_train.jsonl",
256
+ "train_dataset_variant": "full",
257
+ "validation_dataset": "/kaggle/working/datasets/step1_val.jsonl",
258
+ "validation_dataset_variant": "full",
259
+ "validation_split": null,
260
+ "weight_decay": 0.01
261
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:420e24586f4776bba6fefd91d57e1875ddcd3c6040f674ded6c172a1048b1f0a
3
+ size 2799035176