Upload 5 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
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

config.json

@@ -0,0 +1,222 @@
+{
+  "architectures": [
+    "OpenAIPrivacyFilterForTokenClassification"
+  ],
+  "attention_bias": true,
+  "attention_dropout": 0.0,
+  "attention_probs_dropout_prob": null,
+  "bidirectional_context": true,
+  "bidirectional_left_context": null,
+  "bidirectional_right_context": null,
+  "bos_token_id": null,
+  "classifier_dropout": 0.0,
+  "classifier_bias": true,
+  "eos_token_id": 199999,
+  "experts_per_token": null,
+  "head_dim": 64,
+  "hidden_size": 640,
+  "initializer_range": 0.02,
+  "initial_context_length": 4096,
+  "intermediate_size": 640,
+  "layer_norm_eps": null,
+  "max_position_embeddings": 131072,
+  "model_type": "openai_privacy_filter",
+  "num_attention_heads": 14,
+  "num_experts": 128,
+  "num_experts_per_tok": 4,
+  "num_key_value_heads": 2,
+  "num_hidden_layers": 8,
+  "num_local_experts": 128,
+  "output_router_logits": false,
+  "pad_token_id": 199999,
+  "rope_ntk_alpha": null,
+  "rope_ntk_beta": null,
+  "rope_parameters": {
+    "beta_fast": 32.0,
+    "beta_slow": 1.0,
+    "factor": 32.0,
+    "original_max_position_embeddings": 4096,
+    "rope_theta": 150000.0,
+    "rope_type": "yarn",
+    "truncate": false
+  },
+  "rope_scaling_factor": null,
+  "rope_theta": 150000.0,
+  "router_aux_loss_coef": 0.001,
+  "rms_norm_eps": 1e-05,
+  "sliding_window": 128,
+  "swiglu_limit": 7.0,
+  "tie_word_embeddings": false,
+  "use_cache": true,
+  "vocab_size": 200064,
+  "num_labels": 81,
+  "label2id": {
+    "O": 0,
+    "B-private_person": 1,
+    "I-private_person": 2,
+    "E-private_person": 3,
+    "S-private_person": 4,
+    "B-private_organization": 5,
+    "I-private_organization": 6,
+    "E-private_organization": 7,
+    "S-private_organization": 8,
+    "B-private_occupation": 9,
+    "I-private_occupation": 10,
+    "E-private_occupation": 11,
+    "S-private_occupation": 12,
+    "B-private_employee_id": 13,
+    "I-private_employee_id": 14,
+    "E-private_employee_id": 15,
+    "S-private_employee_id": 16,
+    "B-private_employment_status": 17,
+    "I-private_employment_status": 18,
+    "E-private_employment_status": 19,
+    "S-private_employment_status": 20,
+    "B-private_email": 21,
+    "I-private_email": 22,
+    "E-private_email": 23,
+    "S-private_email": 24,
+    "B-private_phone": 25,
+    "I-private_phone": 26,
+    "E-private_phone": 27,
+    "S-private_phone": 28,
+    "B-private_url": 29,
+    "I-private_url": 30,
+    "E-private_url": 31,
+    "S-private_url": 32,
+    "B-private_address": 33,
+    "I-private_address": 34,
+    "E-private_address": 35,
+    "S-private_address": 36,
+    "B-private_location": 37,
+    "I-private_location": 38,
+    "E-private_location": 39,
+    "S-private_location": 40,
+    "B-private_date": 41,
+    "I-private_date": 42,
+    "E-private_date": 43,
+    "S-private_date": 44,
+    "B-private_user_id": 45,
+    "I-private_user_id": 46,
+    "E-private_user_id": 47,
+    "S-private_user_id": 48,
+    "B-private_financial_id": 49,
+    "I-private_financial_id": 50,
+    "E-private_financial_id": 51,
+    "S-private_financial_id": 52,
+    "B-private_government_id": 53,
+    "I-private_government_id": 54,
+    "E-private_government_id": 55,
+    "S-private_government_id": 56,
+    "B-private_health_id": 57,
+    "I-private_health_id": 58,
+    "E-private_health_id": 59,
+    "S-private_health_id": 60,
+    "B-private_device_id": 61,
+    "I-private_device_id": 62,
+    "E-private_device_id": 63,
+    "S-private_device_id": 64,
+    "B-private_network_id": 65,
+    "I-private_network_id": 66,
+    "E-private_network_id": 67,
+    "S-private_network_id": 68,
+    "B-private_vehicle_id": 69,
+    "I-private_vehicle_id": 70,
+    "E-private_vehicle_id": 71,
+    "S-private_vehicle_id": 72,
+    "B-secret": 73,
+    "I-secret": 74,
+    "E-secret": 75,
+    "S-secret": 76,
+    "B-sensitive_attribute": 77,
+    "I-sensitive_attribute": 78,
+    "E-sensitive_attribute": 79,
+    "S-sensitive_attribute": 80
+  },
+  "id2label": {
+    "0": "O",
+    "1": "B-private_person",
+    "2": "I-private_person",
+    "3": "E-private_person",
+    "4": "S-private_person",
+    "5": "B-private_organization",
+    "6": "I-private_organization",
+    "7": "E-private_organization",
+    "8": "S-private_organization",
+    "9": "B-private_occupation",
+    "10": "I-private_occupation",
+    "11": "E-private_occupation",
+    "12": "S-private_occupation",
+    "13": "B-private_employee_id",
+    "14": "I-private_employee_id",
+    "15": "E-private_employee_id",
+    "16": "S-private_employee_id",
+    "17": "B-private_employment_status",
+    "18": "I-private_employment_status",
+    "19": "E-private_employment_status",
+    "20": "S-private_employment_status",
+    "21": "B-private_email",
+    "22": "I-private_email",
+    "23": "E-private_email",
+    "24": "S-private_email",
+    "25": "B-private_phone",
+    "26": "I-private_phone",
+    "27": "E-private_phone",
+    "28": "S-private_phone",
+    "29": "B-private_url",
+    "30": "I-private_url",
+    "31": "E-private_url",
+    "32": "S-private_url",
+    "33": "B-private_address",
+    "34": "I-private_address",
+    "35": "E-private_address",
+    "36": "S-private_address",
+    "37": "B-private_location",
+    "38": "I-private_location",
+    "39": "E-private_location",
+    "40": "S-private_location",
+    "41": "B-private_date",
+    "42": "I-private_date",
+    "43": "E-private_date",
+    "44": "S-private_date",
+    "45": "B-private_user_id",
+    "46": "I-private_user_id",
+    "47": "E-private_user_id",
+    "48": "S-private_user_id",
+    "49": "B-private_financial_id",
+    "50": "I-private_financial_id",
+    "51": "E-private_financial_id",
+    "52": "S-private_financial_id",
+    "53": "B-private_government_id",
+    "54": "I-private_government_id",
+    "55": "E-private_government_id",
+    "56": "S-private_government_id",
+    "57": "B-private_health_id",
+    "58": "I-private_health_id",
+    "59": "E-private_health_id",
+    "60": "S-private_health_id",
+    "61": "B-private_device_id",
+    "62": "I-private_device_id",
+    "63": "E-private_device_id",
+    "64": "S-private_device_id",
+    "65": "B-private_network_id",
+    "66": "I-private_network_id",
+    "67": "E-private_network_id",
+    "68": "S-private_network_id",
+    "69": "B-private_vehicle_id",
+    "70": "I-private_vehicle_id",
+    "71": "E-private_vehicle_id",
+    "72": "S-private_vehicle_id",
+    "73": "B-secret",
+    "74": "I-secret",
+    "75": "E-secret",
+    "76": "S-secret",
+    "77": "B-sensitive_attribute",
+    "78": "I-sensitive_attribute",
+    "79": "E-sensitive_attribute",
+    "80": "S-sensitive_attribute"
+  },
+  "pipeline_config": null,
+  "use_late_interaction": false,
+  "is_regression": null
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04d51c404b03ec68989f581114629e1692d7620d7aa68f65eea869357db12a35
+size 5598086079

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0614fe83cadab421296e664e1f48f4261fa8fef6e03e63bb75c20f38e37d07d3
+size 27868174

tokenizer_config.json

@@ -0,0 +1,13 @@
+{
+  "backend": "tokenizers",
+  "eos_token": "<|endoftext|>",
+  "is_local": true,
+  "local_files_only": false,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 128000,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "TokenizersBackend"
+}

viterbi_calibration.json

@@ -0,0 +1,14 @@
+{
+  "operating_points": {
+    "default": {
+      "biases": {
+        "transition_bias_background_stay": 0.0,
+        "transition_bias_background_to_start": 0.0,
+        "transition_bias_end_to_background": 0.0,
+        "transition_bias_end_to_start": 0.0,
+        "transition_bias_inside_to_continue": 0.0,
+        "transition_bias_inside_to_end": 0.0
+      }
+    }
+  }
+}