Token Classification
Transformers
Safetensors
English
Spanish
qwen3
pii
privacy
named-entity-recognition
text-generation-inference
Instructions to use valendra/privacy-filter-mini with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use valendra/privacy-filter-mini with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="valendra/privacy-filter-mini")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("valendra/privacy-filter-mini") model = AutoModelForTokenClassification.from_pretrained("valendra/privacy-filter-mini", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download evaluation/challenge.json from valendra/privacy-filter-mini: direct link, hf CLI and curl.
- Browser
- Download file 28.2 kB
-
https://huggingface.co/valendra/privacy-filter-mini/resolve/main/evaluation/challenge.json
- Command line
-
hf download hf://valendra/privacy-filter-mini/evaluation/challenge.json
-
curl -L -o challenge.json https://huggingface.co/valendra/privacy-filter-mini/resolve/main/evaluation/challenge.json
28.2 kB
| { | |
| "dataset_manifest_sha256": "9af56087c7655f676c1cb64d214ed2d9d1a5cac78e09aaaee59abd16cd3a0f18", | |
| "split": "challenge", | |
| "documents": 14461, | |
| "models": { | |
| "valendra": { | |
| "metadata": { | |
| "model": "valendra", | |
| "dataset_manifest_sha256": "9af56087c7655f676c1cb64d214ed2d9d1a5cac78e09aaaee59abd16cd3a0f18", | |
| "split": "challenge", | |
| "document_order_sha256": "b7add247b64759824241aee97f400c291c878c1a3a16b8f8358d8fc9b733be84", | |
| "comparison_config_sha256": "808bf293786e733f809daa21722423c7c35719265e0ff977e809d6730f7873bb", | |
| "implementation_sha256": "4cf94a19586c67811e411a011bdfe680e82c1f2c844920dfe2be00ff8080e5f5", | |
| "documents": 14461, | |
| "limit": null, | |
| "sample_seed": null, | |
| "common_categories": [ | |
| "ADDRESS", | |
| "EMAIL", | |
| "PERSON", | |
| "PHONE", | |
| "SECRET", | |
| "URL" | |
| ], | |
| "device": "cuda", | |
| "hardware": "NVIDIA GeForce RTX 4090", | |
| "warmup_documents": 2, | |
| "batch_size": 1, | |
| "scoring": "exact_character_spans_v1", | |
| "model_details": { | |
| "requested_precision": "bf16", | |
| "quantization": "none", | |
| "torch_version": "2.8.0+cu128", | |
| "model_id": "Qwen/Qwen3-0.6B-Base", | |
| "weights_sha256": "130ea7b3a1de096aa5a1c32500d82a8157546c83607d5f71e361df0710fb9d7b", | |
| "parameter_dtypes": { | |
| "torch.bfloat16": 596091945 | |
| }, | |
| "attention_mode": "causal", | |
| "max_length": 512, | |
| "overlap": 64, | |
| "transformers_version": "4.55.4", | |
| "span_scores": true, | |
| "model_revision": "da87bfb608c14b7cf20ba1ce41287e8de496c0cd", | |
| "training_dataset_manifest_sha256": "9af56087c7655f676c1cb64d214ed2d9d1a5cac78e09aaaee59abd16cd3a0f18", | |
| "training_protocol_sha256": "3d55cc7f064fc68f3774d2078d8227b946308ffa4c6242412a83874ee898f073", | |
| "load_seconds": 5.278955338988453, | |
| "label_mapping": { | |
| "PERSON": "PERSON", | |
| "EMAIL": "EMAIL", | |
| "PHONE": "PHONE", | |
| "ADDRESS": "ADDRESS", | |
| "GOVERNMENT_ID": "GOVERNMENT_ID", | |
| "FINANCIAL_ID": "FINANCIAL_ID", | |
| "SECRET": "SECRET", | |
| "IP_ADDRESS": "IP_ADDRESS", | |
| "DATE_OF_BIRTH": "DATE_OF_BIRTH", | |
| "URL": "URL" | |
| } | |
| } | |
| }, | |
| "metrics": { | |
| "documents": 14461, | |
| "untyped": { | |
| "tp": 28169, | |
| "fp": 20817, | |
| "fn": 13721, | |
| "precision": 0.5750418486914629, | |
| "recall": 0.6724516591071855, | |
| "f1": 0.6199436594920551, | |
| "f2": 0.6504160778772179 | |
| }, | |
| "typed_common": { | |
| "tp": 18595, | |
| "fp": 17937, | |
| "fn": 10840, | |
| "precision": 0.5090058031315011, | |
| "recall": 0.6317309325632751, | |
| "f1": 0.5637667318507739, | |
| "f2": 0.6026693113461937 | |
| }, | |
| "common_categories": [ | |
| "ADDRESS", | |
| "EMAIL", | |
| "PERSON", | |
| "PHONE", | |
| "SECRET", | |
| "URL" | |
| ], | |
| "typed_by_category": { | |
| "ADDRESS": { | |
| "tp": 0, | |
| "fp": 8429, | |
| "fn": 1982, | |
| "precision": 0.0, | |
| "recall": 0.0, | |
| "f1": 0.0, | |
| "f2": 0.0 | |
| }, | |
| "EMAIL": { | |
| "tp": 2499, | |
| "fp": 377, | |
| "fn": 0, | |
| "precision": 0.8689151599443672, | |
| "recall": 1.0, | |
| "f1": 0.9298604651162791, | |
| "f2": 0.9707116221255438 | |
| }, | |
| "PERSON": { | |
| "tp": 6427, | |
| "fp": 1317, | |
| "fn": 1050, | |
| "precision": 0.8299328512396694, | |
| "recall": 0.8595693459943827, | |
| "f1": 0.8444911635240785, | |
| "f2": 0.853473919048125 | |
| }, | |
| "PHONE": { | |
| "tp": 2478, | |
| "fp": 6, | |
| "fn": 0, | |
| "precision": 0.9975845410628019, | |
| "recall": 1.0, | |
| "f1": 0.9987908101571946, | |
| "f2": 0.999515972894482 | |
| }, | |
| "SECRET": { | |
| "tp": 4691, | |
| "fp": 2808, | |
| "fn": 2808, | |
| "precision": 0.6255500733431124, | |
| "recall": 0.6255500733431124, | |
| "f1": 0.6255500733431124, | |
| "f2": 0.6255500733431124 | |
| }, | |
| "URL": { | |
| "tp": 2500, | |
| "fp": 5000, | |
| "fn": 5000, | |
| "precision": 0.3333333333333333, | |
| "recall": 0.3333333333333333, | |
| "f1": 0.3333333333333333, | |
| "f2": 0.3333333333333333 | |
| } | |
| }, | |
| "untyped_recall_by_gold_category": { | |
| "ADDRESS": { | |
| "matched": 0, | |
| "total": 1982, | |
| "recall": 0.0 | |
| }, | |
| "DATE_OF_BIRTH": { | |
| "matched": 2500, | |
| "total": 2500, | |
| "recall": 1.0 | |
| }, | |
| "EMAIL": { | |
| "matched": 2499, | |
| "total": 2499, | |
| "recall": 1.0 | |
| }, | |
| "FINANCIAL_ID": { | |
| "matched": 4978, | |
| "total": 4978, | |
| "recall": 1.0 | |
| }, | |
| "GOVERNMENT_ID": { | |
| "matched": 2096, | |
| "total": 2499, | |
| "recall": 0.8387354941976791 | |
| }, | |
| "IP_ADDRESS": { | |
| "matched": 0, | |
| "total": 2478, | |
| "recall": 0.0 | |
| }, | |
| "PERSON": { | |
| "matched": 6427, | |
| "total": 7477, | |
| "recall": 0.8595693459943827 | |
| }, | |
| "PHONE": { | |
| "matched": 2478, | |
| "total": 2478, | |
| "recall": 1.0 | |
| }, | |
| "SECRET": { | |
| "matched": 4691, | |
| "total": 7499, | |
| "recall": 0.6255500733431124 | |
| }, | |
| "URL": { | |
| "matched": 2500, | |
| "total": 7500, | |
| "recall": 0.3333333333333333 | |
| } | |
| }, | |
| "by_language": { | |
| "en": { | |
| "untyped": { | |
| "tp": 12413, | |
| "fp": 6659, | |
| "fn": 3569, | |
| "precision": 0.6508494127516778, | |
| "recall": 0.7766862720560631, | |
| "f1": 0.7082216009585212, | |
| "f2": 0.7477710843373494 | |
| }, | |
| "typed_common": { | |
| "tp": 7413, | |
| "fp": 6658, | |
| "fn": 3569, | |
| "precision": 0.5268282282709118, | |
| "recall": 0.6750136587142597, | |
| "f1": 0.5917854149203688, | |
| "f2": 0.6390627424610767 | |
| } | |
| }, | |
| "es": { | |
| "untyped": { | |
| "tp": 15756, | |
| "fp": 14158, | |
| "fn": 10152, | |
| "precision": 0.526709901718259, | |
| "recall": 0.6081519221861973, | |
| "f1": 0.5645086166744295, | |
| "f2": 0.5899090949934854 | |
| }, | |
| "typed_common": { | |
| "tp": 11182, | |
| "fp": 11279, | |
| "fn": 7271, | |
| "precision": 0.49784070166065625, | |
| "recall": 0.6059719286836829, | |
| "f1": 0.5466099623600724, | |
| "f2": 0.5807443416118745 | |
| } | |
| } | |
| }, | |
| "by_source": { | |
| "synthetic": { | |
| "untyped": { | |
| "tp": 28169, | |
| "fp": 20817, | |
| "fn": 13721, | |
| "precision": 0.5750418486914629, | |
| "recall": 0.6724516591071855, | |
| "f1": 0.6199436594920551, | |
| "f2": 0.6504160778772179 | |
| }, | |
| "typed_common": { | |
| "tp": 18595, | |
| "fp": 17937, | |
| "fn": 10840, | |
| "precision": 0.5090058031315011, | |
| "recall": 0.6317309325632751, | |
| "f1": 0.5637667318507739, | |
| "f2": 0.6026693113461937 | |
| } | |
| } | |
| }, | |
| "typed_excluded_gold": { | |
| "GOVERNMENT_ID": 2499, | |
| "DATE_OF_BIRTH": 2500, | |
| "IP_ADDRESS": 2478, | |
| "FINANCIAL_ID": 4978 | |
| }, | |
| "unmapped_predictions": {}, | |
| "ignored_predictions": 0, | |
| "false_positives_per_document": 1.43952700366503, | |
| "negative_documents": { | |
| "count": 2, | |
| "with_false_positive": 2, | |
| "false_positive_rate": 1.0 | |
| } | |
| }, | |
| "performance": { | |
| "latency_ms": { | |
| "p50": 32.25508285686374, | |
| "p95": 36.32426890544593, | |
| "p99": 38.85766505263746 | |
| }, | |
| "documents_per_second": 30.52227684407281, | |
| "characters_per_second": 3498.987195854014, | |
| "input_tokens_per_second": 1522.7726649514334, | |
| "input_tokens": 721467, | |
| "input_token_definition": "model tokenizer on original text; excludes prompt and repeated windows", | |
| "inference_seconds": 473.7851004325785, | |
| "end_to_end_seconds": 479.30525300791487, | |
| "sampled_peak_rss_bytes": 2073497600, | |
| "peak_vram_allocated_bytes": 1203220480, | |
| "peak_vram_reserved_bytes": 1218445312, | |
| "cpu_process_percent_one_core": 218.9168579762411 | |
| } | |
| }, | |
| "openai": { | |
| "metadata": { | |
| "model": "openai", | |
| "dataset_manifest_sha256": "9af56087c7655f676c1cb64d214ed2d9d1a5cac78e09aaaee59abd16cd3a0f18", | |
| "split": "challenge", | |
| "document_order_sha256": "b7add247b64759824241aee97f400c291c878c1a3a16b8f8358d8fc9b733be84", | |
| "comparison_config_sha256": "808bf293786e733f809daa21722423c7c35719265e0ff977e809d6730f7873bb", | |
| "implementation_sha256": "4cf94a19586c67811e411a011bdfe680e82c1f2c844920dfe2be00ff8080e5f5", | |
| "documents": 14461, | |
| "limit": null, | |
| "sample_seed": null, | |
| "common_categories": [ | |
| "ADDRESS", | |
| "EMAIL", | |
| "PERSON", | |
| "PHONE", | |
| "SECRET", | |
| "URL" | |
| ], | |
| "device": "cuda", | |
| "hardware": "NVIDIA GeForce RTX 4090", | |
| "warmup_documents": 2, | |
| "batch_size": 1, | |
| "scoring": "exact_character_spans_v1", | |
| "model_details": { | |
| "requested_precision": "native", | |
| "quantization": "none", | |
| "torch_version": "2.8.0+cu128", | |
| "model_id": "openai/privacy-filter", | |
| "model_revision": "7ffa9a043d54d1be65afb281eddf0ffbe629385b", | |
| "runtime_revision": "f7f00ca7fb869683eb732c010299d901457f19c3", | |
| "license": "Apache-2.0", | |
| "weights_sha256": "9c262cbe68a0c8a50590a648ef8341a2b7d3be1fa11dfb79893fe0b03ce57b5c", | |
| "parameter_dtypes": { | |
| "torch.bfloat16": 1399475840, | |
| "torch.float32": 10992 | |
| }, | |
| "runtime_context_window": 128000, | |
| "decode_mode": "viterbi", | |
| "calibration_sha256": "bbc8611ef08a55ed72d64856cbbbb9a91db8dfa881f0a92e2afbad6e4bbc775a", | |
| "span_scores": false, | |
| "opf_version": "0.1.0", | |
| "tiktoken_version": "0.11.0", | |
| "load_seconds": 4.347642426146194, | |
| "label_mapping": { | |
| "private_person": "PERSON", | |
| "private_email": "EMAIL", | |
| "private_phone": "PHONE", | |
| "private_address": "ADDRESS", | |
| "private_url": "URL", | |
| "secret": "SECRET", | |
| "private_date": null, | |
| "account_number": null | |
| } | |
| } | |
| }, | |
| "metrics": { | |
| "documents": 14461, | |
| "untyped": { | |
| "tp": 28402, | |
| "fp": 10624, | |
| "fn": 13488, | |
| "precision": 0.7277712294367857, | |
| "recall": 0.6780138457865839, | |
| "f1": 0.7020119630233822, | |
| "f2": 0.6874134742915783 | |
| }, | |
| "typed_common": { | |
| "tp": 16642, | |
| "fp": 10673, | |
| "fn": 12793, | |
| "precision": 0.6092623100860333, | |
| "recall": 0.5653813487344997, | |
| "f1": 0.5865022026431719, | |
| "f2": 0.5736444796801213 | |
| }, | |
| "common_categories": [ | |
| "ADDRESS", | |
| "EMAIL", | |
| "PERSON", | |
| "PHONE", | |
| "SECRET", | |
| "URL" | |
| ], | |
| "typed_by_category": { | |
| "ADDRESS": { | |
| "tp": 187, | |
| "fp": 5517, | |
| "fn": 1795, | |
| "precision": 0.032784011220196356, | |
| "recall": 0.09434914228052473, | |
| "f1": 0.04865990111891751, | |
| "f2": 0.06858861502347419 | |
| }, | |
| "EMAIL": { | |
| "tp": 2444, | |
| "fp": 33, | |
| "fn": 55, | |
| "precision": 0.9866774323778764, | |
| "recall": 0.9779911964785915, | |
| "f1": 0.9823151125401929, | |
| "f2": 0.9797161869638419 | |
| }, | |
| "PERSON": { | |
| "tp": 6602, | |
| "fp": 1081, | |
| "fn": 875, | |
| "precision": 0.859299752700768, | |
| "recall": 0.882974454995319, | |
| "f1": 0.8709762532981531, | |
| "f2": 0.8781357239764838 | |
| }, | |
| "PHONE": { | |
| "tp": 2419, | |
| "fp": 64, | |
| "fn": 59, | |
| "precision": 0.9742247281514297, | |
| "recall": 0.9761904761904762, | |
| "f1": 0.9752066115702479, | |
| "f2": 0.9757966922146027 | |
| }, | |
| "SECRET": { | |
| "tp": 11, | |
| "fp": 568, | |
| "fn": 7488, | |
| "precision": 0.018998272884283247, | |
| "recall": 0.0014668622482997733, | |
| "f1": 0.0027234463976231744, | |
| "f2": 0.0017988552739165984 | |
| }, | |
| "URL": { | |
| "tp": 4979, | |
| "fp": 3410, | |
| "fn": 2521, | |
| "precision": 0.5935153176779115, | |
| "recall": 0.6638666666666667, | |
| "f1": 0.6267228900497198, | |
| "f2": 0.648493057907213 | |
| } | |
| }, | |
| "untyped_recall_by_gold_category": { | |
| "ADDRESS": { | |
| "matched": 187, | |
| "total": 1982, | |
| "recall": 0.09434914228052473 | |
| }, | |
| "DATE_OF_BIRTH": { | |
| "matched": 2500, | |
| "total": 2500, | |
| "recall": 1.0 | |
| }, | |
| "EMAIL": { | |
| "matched": 2444, | |
| "total": 2499, | |
| "recall": 0.9779911964785915 | |
| }, | |
| "FINANCIAL_ID": { | |
| "matched": 4655, | |
| "total": 4978, | |
| "recall": 0.9351145038167938 | |
| }, | |
| "GOVERNMENT_ID": { | |
| "matched": 2421, | |
| "total": 2499, | |
| "recall": 0.9687875150060024 | |
| }, | |
| "IP_ADDRESS": { | |
| "matched": 326, | |
| "total": 2478, | |
| "recall": 0.13155770782889428 | |
| }, | |
| "PERSON": { | |
| "matched": 6634, | |
| "total": 7477, | |
| "recall": 0.8872542463554902 | |
| }, | |
| "PHONE": { | |
| "matched": 2419, | |
| "total": 2478, | |
| "recall": 0.9761904761904762 | |
| }, | |
| "SECRET": { | |
| "matched": 1837, | |
| "total": 7499, | |
| "recall": 0.24496599546606215 | |
| }, | |
| "URL": { | |
| "matched": 4979, | |
| "total": 7500, | |
| "recall": 0.6638666666666667 | |
| } | |
| }, | |
| "by_language": { | |
| "en": { | |
| "untyped": { | |
| "tp": 9841, | |
| "fp": 5819, | |
| "fn": 6141, | |
| "precision": 0.6284163473818646, | |
| "recall": 0.6157552246277062, | |
| "f1": 0.6220213640098603, | |
| "f2": 0.6182464693169825 | |
| }, | |
| "typed_common": { | |
| "tp": 5139, | |
| "fp": 5761, | |
| "fn": 5843, | |
| "precision": 0.4714678899082569, | |
| "recall": 0.46794755053724274, | |
| "f1": 0.4697011242116808, | |
| "f2": 0.4686474064346684 | |
| } | |
| }, | |
| "es": { | |
| "untyped": { | |
| "tp": 18561, | |
| "fp": 4805, | |
| "fn": 7347, | |
| "precision": 0.7943593255157065, | |
| "recall": 0.7164196387216304, | |
| "f1": 0.7533790640094168, | |
| "f2": 0.7307595395203074 | |
| }, | |
| "typed_common": { | |
| "tp": 11503, | |
| "fp": 4912, | |
| "fn": 6950, | |
| "precision": 0.7007614986293025, | |
| "recall": 0.6233674741234487, | |
| "f1": 0.659802684409774, | |
| "f2": 0.6374477706229841 | |
| } | |
| } | |
| }, | |
| "by_source": { | |
| "synthetic": { | |
| "untyped": { | |
| "tp": 28402, | |
| "fp": 10624, | |
| "fn": 13488, | |
| "precision": 0.7277712294367857, | |
| "recall": 0.6780138457865839, | |
| "f1": 0.7020119630233822, | |
| "f2": 0.6874134742915783 | |
| }, | |
| "typed_common": { | |
| "tp": 16642, | |
| "fp": 10673, | |
| "fn": 12793, | |
| "precision": 0.6092623100860333, | |
| "recall": 0.5653813487344997, | |
| "f1": 0.5865022026431719, | |
| "f2": 0.5736444796801213 | |
| } | |
| } | |
| }, | |
| "typed_excluded_gold": { | |
| "GOVERNMENT_ID": 2499, | |
| "DATE_OF_BIRTH": 2500, | |
| "IP_ADDRESS": 2478, | |
| "FINANCIAL_ID": 4978 | |
| }, | |
| "unmapped_predictions": { | |
| "account_number": 9206, | |
| "private_date": 2505 | |
| }, | |
| "ignored_predictions": 0, | |
| "false_positives_per_document": 0.7346656524445059, | |
| "negative_documents": { | |
| "count": 2, | |
| "with_false_positive": 2, | |
| "false_positive_rate": 1.0 | |
| } | |
| }, | |
| "performance": { | |
| "latency_ms": { | |
| "p50": 18.598005874082446, | |
| "p95": 20.177082857117057, | |
| "p99": 22.28398541919887 | |
| }, | |
| "documents_per_second": 53.1927473556665, | |
| "characters_per_second": 6097.86559700633, | |
| "input_tokens_per_second": 1904.7829560142147, | |
| "input_tokens": 517835, | |
| "input_token_definition": "model tokenizer on original text; excludes prompt and repeated windows", | |
| "inference_seconds": 271.8603704243433, | |
| "end_to_end_seconds": 275.67499247402884, | |
| "sampled_peak_rss_bytes": 1649799168, | |
| "peak_vram_allocated_bytes": 3080668160, | |
| "peak_vram_reserved_bytes": 3332374528, | |
| "cpu_process_percent_one_core": 274.424601669762 | |
| } | |
| }, | |
| "gliner": { | |
| "metadata": { | |
| "model": "gliner", | |
| "dataset_manifest_sha256": "9af56087c7655f676c1cb64d214ed2d9d1a5cac78e09aaaee59abd16cd3a0f18", | |
| "split": "challenge", | |
| "document_order_sha256": "b7add247b64759824241aee97f400c291c878c1a3a16b8f8358d8fc9b733be84", | |
| "comparison_config_sha256": "808bf293786e733f809daa21722423c7c35719265e0ff977e809d6730f7873bb", | |
| "implementation_sha256": "4cf94a19586c67811e411a011bdfe680e82c1f2c844920dfe2be00ff8080e5f5", | |
| "documents": 14461, | |
| "limit": null, | |
| "sample_seed": null, | |
| "common_categories": [ | |
| "ADDRESS", | |
| "EMAIL", | |
| "PERSON", | |
| "PHONE", | |
| "SECRET", | |
| "URL" | |
| ], | |
| "device": "cuda", | |
| "hardware": "NVIDIA GeForce RTX 4090", | |
| "warmup_documents": 2, | |
| "batch_size": 1, | |
| "scoring": "exact_character_spans_v1", | |
| "model_details": { | |
| "requested_precision": "bf16", | |
| "quantization": "none", | |
| "torch_version": "2.8.0+cu128", | |
| "model_id": "nvidia/gliner-PII", | |
| "model_revision": "bd23e8ef4425fd04e34c5204ab49ffaa706eae79", | |
| "license": "NVIDIA-Open-Model-License", | |
| "weights_sha256": "a4dfd0dcbd718acc86dca65fa99f1097d2796c8d7a681e1bc42f40f946c03802", | |
| "parameter_dtypes": { | |
| "torch.bfloat16": 445463040 | |
| }, | |
| "gliner_version": "0.2.24", | |
| "transformers_version": "4.57.3", | |
| "label_prompts": [ | |
| "first_name", | |
| "last_name", | |
| "name", | |
| "email", | |
| "phone_number", | |
| "street_address", | |
| "address", | |
| "url", | |
| "api_key", | |
| "password", | |
| "http_cookie", | |
| "cvv", | |
| "pin", | |
| "national_id", | |
| "ssn", | |
| "tax_id", | |
| "account_number", | |
| "bank_routing_number", | |
| "credit_debit_card", | |
| "swift_bic", | |
| "date_of_birth", | |
| "ipv4", | |
| "ipv6" | |
| ], | |
| "threshold": 0.3, | |
| "overlap_words": 32, | |
| "max_words": 384, | |
| "max_encoder_tokens": 512, | |
| "span_scores": true, | |
| "load_seconds": 11.4082525849808, | |
| "label_mapping": { | |
| "first_name": "PERSON", | |
| "last_name": "PERSON", | |
| "name": "PERSON", | |
| "email": "EMAIL", | |
| "phone_number": "PHONE", | |
| "street_address": "ADDRESS", | |
| "address": "ADDRESS", | |
| "url": "URL", | |
| "api_key": "SECRET", | |
| "password": "SECRET", | |
| "http_cookie": "SECRET", | |
| "cvv": "SECRET", | |
| "pin": "SECRET", | |
| "national_id": "GOVERNMENT_ID", | |
| "ssn": "GOVERNMENT_ID", | |
| "tax_id": "GOVERNMENT_ID", | |
| "account_number": "FINANCIAL_ID", | |
| "bank_routing_number": "FINANCIAL_ID", | |
| "credit_debit_card": "FINANCIAL_ID", | |
| "swift_bic": "FINANCIAL_ID", | |
| "date_of_birth": "DATE_OF_BIRTH", | |
| "ipv4": "IP_ADDRESS", | |
| "ipv6": "IP_ADDRESS" | |
| } | |
| } | |
| }, | |
| "metrics": { | |
| "documents": 14461, | |
| "untyped": { | |
| "tp": 15238, | |
| "fp": 24262, | |
| "fn": 26652, | |
| "precision": 0.38577215189873415, | |
| "recall": 0.3637622344234901, | |
| "f1": 0.3744440348937216, | |
| "f2": 0.3679609774944461 | |
| }, | |
| "typed_common": { | |
| "tp": 7790, | |
| "fp": 26030, | |
| "fn": 21645, | |
| "precision": 0.2303370786516854, | |
| "recall": 0.2646509257686428, | |
| "f1": 0.24630463994941112, | |
| "f2": 0.2569939297967802 | |
| }, | |
| "common_categories": [ | |
| "ADDRESS", | |
| "EMAIL", | |
| "PERSON", | |
| "PHONE", | |
| "SECRET", | |
| "URL" | |
| ], | |
| "typed_by_category": { | |
| "ADDRESS": { | |
| "tp": 286, | |
| "fp": 1621, | |
| "fn": 1696, | |
| "precision": 0.14997378080755114, | |
| "recall": 0.1442986881937437, | |
| "f1": 0.14708151195680125, | |
| "f2": 0.14539908490086428 | |
| }, | |
| "EMAIL": { | |
| "tp": 2499, | |
| "fp": 1, | |
| "fn": 0, | |
| "precision": 0.9996, | |
| "recall": 1.0, | |
| "f1": 0.9997999599919984, | |
| "f2": 0.9999199743918054 | |
| }, | |
| "PERSON": { | |
| "tp": 6, | |
| "fp": 17470, | |
| "fn": 7471, | |
| "precision": 0.0003433279926756695, | |
| "recall": 0.0008024608800320985, | |
| "f1": 0.00048090409970745, | |
| "f2": 0.0006331251055208511 | |
| }, | |
| "PHONE": { | |
| "tp": 0, | |
| "fp": 1, | |
| "fn": 2478, | |
| "precision": 0.0, | |
| "recall": 0.0, | |
| "f1": 0.0, | |
| "f2": 0.0 | |
| }, | |
| "SECRET": { | |
| "tp": 2499, | |
| "fp": 6937, | |
| "fn": 5000, | |
| "precision": 0.26483679525222553, | |
| "recall": 0.3332444325910121, | |
| "f1": 0.2951284322409211, | |
| "f2": 0.31687461959829577 | |
| }, | |
| "URL": { | |
| "tp": 2500, | |
| "fp": 0, | |
| "fn": 5000, | |
| "precision": 1.0, | |
| "recall": 0.3333333333333333, | |
| "f1": 0.5, | |
| "f2": 0.3846153846153846 | |
| } | |
| }, | |
| "untyped_recall_by_gold_category": { | |
| "ADDRESS": { | |
| "matched": 286, | |
| "total": 1982, | |
| "recall": 0.1442986881937437 | |
| }, | |
| "DATE_OF_BIRTH": { | |
| "matched": 2500, | |
| "total": 2500, | |
| "recall": 1.0 | |
| }, | |
| "EMAIL": { | |
| "matched": 2499, | |
| "total": 2499, | |
| "recall": 1.0 | |
| }, | |
| "FINANCIAL_ID": { | |
| "matched": 2498, | |
| "total": 4978, | |
| "recall": 0.5018079550020088 | |
| }, | |
| "GOVERNMENT_ID": { | |
| "matched": 2449, | |
| "total": 2499, | |
| "recall": 0.9799919967987195 | |
| }, | |
| "IP_ADDRESS": { | |
| "matched": 1, | |
| "total": 2478, | |
| "recall": 0.0004035512510088781 | |
| }, | |
| "PERSON": { | |
| "matched": 6, | |
| "total": 7477, | |
| "recall": 0.0008024608800320985 | |
| }, | |
| "PHONE": { | |
| "matched": 0, | |
| "total": 2478, | |
| "recall": 0.0 | |
| }, | |
| "SECRET": { | |
| "matched": 2499, | |
| "total": 7499, | |
| "recall": 0.3332444325910121 | |
| }, | |
| "URL": { | |
| "matched": 2500, | |
| "total": 7500, | |
| "recall": 0.3333333333333333 | |
| } | |
| }, | |
| "by_language": { | |
| "en": { | |
| "untyped": { | |
| "tp": 7784, | |
| "fp": 11083, | |
| "fn": 8198, | |
| "precision": 0.4125722160385859, | |
| "recall": 0.48704792892003507, | |
| "f1": 0.446727309248472, | |
| "f2": 0.47007669545262404 | |
| }, | |
| "typed_common": { | |
| "tp": 2786, | |
| "fp": 11077, | |
| "fn": 8196, | |
| "precision": 0.2009666017456539, | |
| "recall": 0.25368785285011836, | |
| "f1": 0.22427047695713423, | |
| "f2": 0.24104099254209135 | |
| } | |
| }, | |
| "es": { | |
| "untyped": { | |
| "tp": 7454, | |
| "fp": 13179, | |
| "fn": 18454, | |
| "precision": 0.36126593321378375, | |
| "recall": 0.287710359734445, | |
| "f1": 0.32031971809802107, | |
| "f2": 0.29992355047680364 | |
| }, | |
| "typed_common": { | |
| "tp": 5004, | |
| "fp": 14953, | |
| "fn": 13449, | |
| "precision": 0.2507390890414391, | |
| "recall": 0.2711754186311169, | |
| "f1": 0.26055714657641244, | |
| "f2": 0.2668259232795487 | |
| } | |
| } | |
| }, | |
| "by_source": { | |
| "synthetic": { | |
| "untyped": { | |
| "tp": 15238, | |
| "fp": 24262, | |
| "fn": 26652, | |
| "precision": 0.38577215189873415, | |
| "recall": 0.3637622344234901, | |
| "f1": 0.3744440348937216, | |
| "f2": 0.3679609774944461 | |
| }, | |
| "typed_common": { | |
| "tp": 7790, | |
| "fp": 26030, | |
| "fn": 21645, | |
| "precision": 0.2303370786516854, | |
| "recall": 0.2646509257686428, | |
| "f1": 0.24630463994941112, | |
| "f2": 0.2569939297967802 | |
| } | |
| } | |
| }, | |
| "typed_excluded_gold": { | |
| "GOVERNMENT_ID": 2499, | |
| "DATE_OF_BIRTH": 2500, | |
| "FINANCIAL_ID": 4978, | |
| "IP_ADDRESS": 2478 | |
| }, | |
| "unmapped_predictions": {}, | |
| "ignored_predictions": 0, | |
| "false_positives_per_document": 1.6777539589240025, | |
| "negative_documents": { | |
| "count": 2, | |
| "with_false_positive": 2, | |
| "false_positive_rate": 1.0 | |
| } | |
| }, | |
| "performance": { | |
| "latency_ms": { | |
| "p50": 40.188657119870186, | |
| "p95": 43.96042600274086, | |
| "p99": 46.425058972090476 | |
| }, | |
| "documents_per_second": 24.748330706898397, | |
| "characters_per_second": 2837.078397020506, | |
| "input_tokens_per_second": 1000.718373156295, | |
| "input_tokens": 584742, | |
| "input_token_definition": "model tokenizer on original text; excludes prompt and repeated windows", | |
| "inference_seconds": 584.3222385891713, | |
| "end_to_end_seconds": 590.5962648310233, | |
| "sampled_peak_rss_bytes": 2059366400, | |
| "peak_vram_allocated_bytes": 915900928, | |
| "peak_vram_reserved_bytes": 937426944, | |
| "cpu_process_percent_one_core": 100.08867905202999 | |
| } | |
| } | |
| } | |
| } |