Text Classification
Transformers
Safetensors
English
sentinel_stage_a
feature-extraction
custom
compliance
finance
risk-detection
sentinel-stage-a
limited-functionality
model-version:sentinel-mb-c-d11-20260424
custom_code
Instructions to use AurelexAI/sentinel-01-pub with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AurelexAI/sentinel-01-pub with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="AurelexAI/sentinel-01-pub", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("AurelexAI/sentinel-01-pub", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "architectures": [ | |
| "SentinelStageAModel" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "configuration_sentinel.SentinelConfig", | |
| "AutoModel": "modeling_sentinel.SentinelStageAModel" | |
| }, | |
| "classifier_dropout": 0.1, | |
| "custom_pipelines": { | |
| "sentinel-stage-a": { | |
| "impl": "pipeline_sentinel.SentinelStageAPipeline", | |
| "pt": [ | |
| "AutoModel" | |
| ], | |
| "type": "text" | |
| } | |
| }, | |
| "dataset_signature": { | |
| "counts": { | |
| "dev": 150, | |
| "test": 150, | |
| "train": 900 | |
| }, | |
| "distribution": { | |
| "dev": { | |
| "clean": 8, | |
| "risky": 142 | |
| }, | |
| "test": { | |
| "clean": 8, | |
| "risky": 142 | |
| }, | |
| "train": { | |
| "clean": 297, | |
| "risky": 603 | |
| } | |
| }, | |
| "generator_version": "2026-04-07-final-audit-clear-v1" | |
| }, | |
| "encoder_code_revision": null, | |
| "encoder_config": { | |
| "_attn_implementation_autoset": true, | |
| "_name_or_path": "answerdotai/ModernBERT-base", | |
| "add_cross_attention": false, | |
| "architectures": [ | |
| "ModernBertForMaskedLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bad_words_ids": null, | |
| "begin_suppress_tokens": null, | |
| "bos_token_id": 50281, | |
| "chunk_size_feed_forward": 0, | |
| "classifier_activation": "gelu", | |
| "classifier_bias": false, | |
| "classifier_dropout": 0.0, | |
| "classifier_pooling": "mean", | |
| "cls_token_id": 50281, | |
| "cross_attention_hidden_size": null, | |
| "decoder_bias": true, | |
| "decoder_start_token_id": null, | |
| "deterministic_flash_attn": false, | |
| "diversity_penalty": 0.0, | |
| "do_sample": false, | |
| "early_stopping": false, | |
| "embedding_dropout": 0.0, | |
| "encoder_no_repeat_ngram_size": 0, | |
| "eos_token_id": 50282, | |
| "exponential_decay_length_penalty": null, | |
| "finetuning_task": null, | |
| "forced_bos_token_id": null, | |
| "forced_eos_token_id": null, | |
| "global_attn_every_n_layers": 3, | |
| "global_rope_theta": 160000.0, | |
| "gradient_checkpointing": false, | |
| "hidden_activation": "gelu", | |
| "hidden_size": 768, | |
| "id2label": { | |
| "0": "LABEL_0", | |
| "1": "LABEL_1" | |
| }, | |
| "initializer_cutoff_factor": 2.0, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 1152, | |
| "is_decoder": false, | |
| "is_encoder_decoder": false, | |
| "label2id": { | |
| "LABEL_0": 0, | |
| "LABEL_1": 1 | |
| }, | |
| "layer_norm_eps": 1e-05, | |
| "length_penalty": 1.0, | |
| "local_attention": 128, | |
| "local_rope_theta": 10000.0, | |
| "max_length": 20, | |
| "max_position_embeddings": 8192, | |
| "min_length": 0, | |
| "mlp_bias": false, | |
| "mlp_dropout": 0.0, | |
| "model_type": "modernbert", | |
| "no_repeat_ngram_size": 0, | |
| "norm_bias": false, | |
| "norm_eps": 1e-05, | |
| "num_attention_heads": 12, | |
| "num_beam_groups": 1, | |
| "num_beams": 1, | |
| "num_hidden_layers": 22, | |
| "num_return_sequences": 1, | |
| "output_attentions": false, | |
| "output_hidden_states": false, | |
| "output_scores": false, | |
| "pad_token_id": 50283, | |
| "position_embedding_type": "absolute", | |
| "prefix": null, | |
| "problem_type": null, | |
| "pruned_heads": {}, | |
| "reference_compile": null, | |
| "remove_invalid_values": false, | |
| "repad_logits_with_grad": false, | |
| "repetition_penalty": 1.0, | |
| "return_dict": true, | |
| "return_dict_in_generate": false, | |
| "sep_token_id": 50282, | |
| "sparse_pred_ignore_index": -100, | |
| "sparse_prediction": false, | |
| "suppress_tokens": null, | |
| "task_specific_params": null, | |
| "temperature": 1.0, | |
| "tf_legacy_loss": false, | |
| "tie_encoder_decoder": false, | |
| "tie_word_embeddings": true, | |
| "tokenizer_class": null, | |
| "top_k": 50, | |
| "top_p": 1.0, | |
| "torch_dtype": "float32", | |
| "torchscript": false, | |
| "transformers_version": "4.48.3", | |
| "typical_p": 1.0, | |
| "use_bfloat16": false, | |
| "vocab_size": 50368 | |
| }, | |
| "encoder_config_overrides": {}, | |
| "encoder_model_name": "answerdotai/ModernBERT-base", | |
| "encoder_revision": null, | |
| "encoder_trust_remote_code": false, | |
| "head_code": "c", | |
| "head_div": 1, | |
| "head_dropout": 0.1, | |
| "head_mul": 1, | |
| "head_skip": true, | |
| "head_type": "columnar", | |
| "head_variant": "d11", | |
| "max_length": 512, | |
| "model_key": "sentinel-mb-c-d11", | |
| "model_type": "sentinel_stage_a", | |
| "model_version": "sentinel-mb-c-d11-20260424", | |
| "output_heads": [ | |
| "violation", | |
| "severity", | |
| "domain", | |
| "subtype", | |
| "jurisdiction", | |
| "why", | |
| "impacted_principles", | |
| "remediation_actions", | |
| "content_type", | |
| "audience_segment", | |
| "detection_difficulty", | |
| "aggravating_factors" | |
| ], | |
| "output_signature": { | |
| "aggravating_factors": { | |
| "labels": [ | |
| "intentional", | |
| "reckless", | |
| "negligent", | |
| "concealment_present", | |
| "customer_harm_potential", | |
| "financial_benefit_to_respondent", | |
| "vulnerable_client", | |
| "pattern_or_duration" | |
| ], | |
| "type": "multilabel" | |
| }, | |
| "audience_segment": { | |
| "labels": [ | |
| "client", | |
| "internal", | |
| "prospect_or_investor", | |
| "public", | |
| "third_party" | |
| ], | |
| "type": "multiclass" | |
| }, | |
| "content_type": { | |
| "labels": [ | |
| "email", | |
| "message" | |
| ], | |
| "type": "multiclass" | |
| }, | |
| "detection_difficulty": { | |
| "labels": [ | |
| "obvious", | |
| "moderate", | |
| "subtle" | |
| ], | |
| "type": "multiclass" | |
| }, | |
| "domain": { | |
| "labels": [ | |
| "performance_claims_forecasting", | |
| "investment_advice_suitability", | |
| "conflicts_inducements", | |
| "marketing_solicitation_advertising", | |
| "selective_disclosure_fair_access", | |
| "mnpi_insider_trading", | |
| "recordkeeping_supervision", | |
| "ai_automation_capability_claims", | |
| "privacy_confidentiality", | |
| "cybersecurity_internal_controls", | |
| "employment_favoritism_role_conflict", | |
| "aml_and_suspicious_activity", | |
| "other_unknown" | |
| ], | |
| "type": "multiclass" | |
| }, | |
| "impacted_principles": { | |
| "labels": [ | |
| "truthful_non_misleading_communications", | |
| "balanced_risk_reward_presentation", | |
| "no_performance_guarantees_or_promissory_language", | |
| "registration_and_scope_of_advice", | |
| "duty_of_loyalty_conflict_disclosure", | |
| "fair_access_to_material_information", | |
| "insider_trading_and_mnpi_controls", | |
| "supervision_and_books_records", | |
| "privacy_confidentiality_and_secure_handling", | |
| "security_control_integrity", | |
| "role_separation_and_fair_access_in_academia", | |
| "non_coercion_and_no_undue_influence", | |
| "accurate_ai_capability_and_human_oversight", | |
| "client_vulnerability_and_exploitation_prevention", | |
| "aml_and_sanctions_compliance" | |
| ], | |
| "type": "multilabel" | |
| }, | |
| "jurisdiction": { | |
| "labels": [ | |
| "US", | |
| "EU", | |
| "UK", | |
| "Other", | |
| "Unknown" | |
| ], | |
| "type": "multiclass" | |
| }, | |
| "remediation_actions": { | |
| "labels": [ | |
| "add_forward_looking_disclaimer", | |
| "reframe_as_scenarios_not_expectations", | |
| "add_balanced_risk_and_downside_section", | |
| "remove_or_soften_guarantee_language", | |
| "remove_personalized_recommendations", | |
| "add_registered_advice_boundary_language", | |
| "disclose_conflicts_and_compensation", | |
| "add_fees_costs_and_alternatives_comparison", | |
| "use_standardized_approved_performance_materials", | |
| "add_performance_methodology_and_gross_net_context", | |
| "avoid_selective_disclosure_share_broadly", | |
| "escalate_mnpi_to_compliance_and_halt", | |
| "keep_discussion_on_retained_channels", | |
| "require_formal_preapproval_before_send", | |
| "remove_pressure_scarcity_and_use_factual_timeline", | |
| "substantiation_or_remove_credibility_claims", | |
| "add_testimonial_endorsement_and_rating_disclosure", | |
| "make_required_disclosure_clear_and_prominent", | |
| "avoid_minimizing_compliance_or_diligence", | |
| "clarify_ai_is_assistive_with_human_review", | |
| "remove_claims_that_ai_eliminates_risk", | |
| "redact_and_minimize_sensitive_data", | |
| "use_secure_transfer_and_limit_access", | |
| "avoid_sharing_internal_controls_or_sanitize", | |
| "route_academic_opportunities_through_institution", | |
| "separate_recommendation_letters_from_work", | |
| "assess_cost_to_equity_against_client_profile", | |
| "flag_for_elder_exploitation_review_and_hold", | |
| "assess_sar_filing_obligation_and_escalate", | |
| "initiate_breach_notification_review_and_timeline", | |
| "remove_provisions_impeding_regulatory_communications" | |
| ], | |
| "type": "multilabel" | |
| }, | |
| "severity": { | |
| "labels": [ | |
| "sev_0_compliant_or_ok", | |
| "sev_1_minor", | |
| "sev_2_moderate", | |
| "sev_3_high" | |
| ], | |
| "type": "multiclass" | |
| }, | |
| "subtype": { | |
| "labels": [ | |
| "speculative_outcomes_unqualified", | |
| "implicit_or_explicit_guarantee", | |
| "risk_context_omitted_or_unbalanced", | |
| "unregistered_personalized_investment_advice", | |
| "undisclosed_economic_conflict_or_referral", | |
| "pressure_or_coercion", | |
| "selective_disclosure", | |
| "mnpi_misuse_or_encouragement", | |
| "recordkeeping_or_preapproval_evasion", | |
| "ai_autonomy_or_safety_overstatement", | |
| "credentials_validation_or_compliance_misrepresentation", | |
| "confidential_data_leakage", | |
| "internal_controls_or_exception_process_leakage", | |
| "academic_commercial_role_blurring_or_quid_pro_quo", | |
| "improper_solicitation_offering_pressure", | |
| "excessive_trading_or_account_churning", | |
| "product_switching_without_cost_benefit_analysis", | |
| "dual_registrant_capacity_or_wrap_fee_conflict_confusion", | |
| "elder_exploitation_or_vulnerable_client_signal", | |
| "suspicious_activity_indicator_or_structuring", | |
| "influencer_or_social_media_promotion_compliance_failure", | |
| "crypto_asset_misrepresentation_or_inadequate_disclosure", | |
| "other_unknown" | |
| ], | |
| "type": "multiclass" | |
| }, | |
| "violation": { | |
| "type": "binary" | |
| }, | |
| "why": { | |
| "labels": [ | |
| "forward_looking_statement_unqualified", | |
| "guarantee_or_assurance_language", | |
| "omits_material_risk_or_downside", | |
| "implies_downside_protection_or_no_drawdown", | |
| "cherry_picks_performance_period", | |
| "omits_performance_methodology_or_gross_net_context", | |
| "personalized_trade_or_allocation_recommendation", | |
| "timing_or_sizing_guidance", | |
| "creates_implied_advisory_relationship", | |
| "conflict_not_disclosed", | |
| "referral_relationship_not_disclosed", | |
| "omits_fees_costs_or_reasonably_available_alternatives", | |
| "selective_private_performance_or_fundraising_update", | |
| "off_the_record_or_not_in_writing_language", | |
| "mnpi_possession_indicated", | |
| "encourages_action_before_public_release", | |
| "avoid_recordkeeping_channel_shift", | |
| "bypasses_required_preapproval", | |
| "pressure_scarcity_urgency", | |
| "unsubstantiated_social_proof_or_validation", | |
| "omits_testimonial_endorsement_or_rating_disclosure", | |
| "obscures_required_disclosure_or_form_crs", | |
| "minimizes_need_for_diligence_or_compliance", | |
| "overstates_ai_capability_or_removes_human_oversight", | |
| "claims_compliance_risk_eliminated", | |
| "shares_sensitive_personal_or_financial_data", | |
| "violates_need_to_know_data_minimization", | |
| "shares_sensitive_internal_controls_or_exceptions", | |
| "role_power_imbalance_or_favoritism", | |
| "excessive_trading_cost_to_equity", | |
| "inadequate_customer_profile_or_suitability_basis", | |
| "exploits_vulnerable_or_elderly_client", | |
| "aml_suspicious_activity_indicator", | |
| "omits_switching_costs_and_product_comparison", | |
| "conflict_language_understates_actual_relationship", | |
| "omits_influencer_compensation_or_affiliation_disclosure", | |
| "misrepresents_sipc_or_regulatory_protection_for_crypto", | |
| "data_breach_notification_obligation_triggered", | |
| "impedes_regulatory_reporting_or_whistleblower_rights" | |
| ], | |
| "type": "multilabel" | |
| } | |
| }, | |
| "projection_size": 640, | |
| "release_alias_of": null, | |
| "release_channel": "sentinel-01-pub", | |
| "release_repo_id": "AurelexAI/sentinel-01-pub", | |
| "thresholds": { | |
| "aggravating_factors": 0.4, | |
| "impacted_principles": 0.7, | |
| "remediation_actions": 0.5, | |
| "violation": 0.5, | |
| "why": 0.55 | |
| }, | |
| "tokenizer_class": "PreTrainedTokenizerFast", | |
| "torch_dtype": "float32", | |
| "trainable_head_params": 14325653, | |
| "transformers_version": "4.48.3" | |
| } | |