MCPcopy Create free account

hub / github.com/elder-plinius/OBLITERATUS / functions

Functions1,841 in github.com/elder-plinius/OBLITERATUS

Methodtest_direction_specificity_test
(self)
tests/test_module_imports.py:50
Methodtest_direction_stats
(self)
tests/test_telemetry.py:157
Methodtest_directions_are_unit_vectors
Per-expert directions should be unit normalized.
tests/test_abliterate.py:2163
Methodtest_directions_are_unit_vectors
Extracted directions should be unit length.
tests/test_analysis.py:40
Methodtest_disable_programmatically
(self)
tests/test_telemetry.py:79
Methodtest_disable_via_env_false
(self)
tests/test_telemetry.py:65
Methodtest_disable_via_env_zero
(self)
tests/test_telemetry.py:60
Methodtest_disabled_by_default
(self)
tests/test_telemetry.py:46
Methodtest_distributed_detects_not_sft
Distributed refusal should not be predicted as SFT.
tests/test_novel_analysis.py:249
Methodtest_does_not_send_when_disabled
(self)
tests/test_telemetry.py:328
Methodtest_does_nothing_when_disabled
(self)
tests/test_telemetry.py:421
Methodtest_does_nothing_when_disabled
(self)
tests/test_telemetry.py:521
Methodtest_double_projection_is_idempotent
Projecting twice should give the same result as projecting once.
tests/test_abliteration_math.py:86
Methodtest_dpo_zero_regularization
(self)
tests/test_informed_pipeline.py:177
Methodtest_effective_rank
Multi-direction subspace should yield effective rank > 1.
tests/test_telemetry.py:235
Methodtest_effective_rank_change
Should compute before/after rank comparison.
tests/test_advanced_metrics.py:180
Methodtest_effective_rank_nan_matrix
effective_rank should handle matrix with NaN.
tests/test_edge_cases.py:111
Methodtest_ega_distill_integration
EGA should be called during distill when per_expert_directions is enabled.
tests/test_abliterate.py:2553
Methodtest_ega_only_on_primary_direction
EGA should only apply for dir_idx==0, not higher SVD directions.
tests/test_abliterate.py:2543
Methodtest_ega_pipeline_flags
Pipeline with surgical method should enable per_expert_directions.
tests/test_abliterate.py:2538
Methodtest_ega_with_routing_data_end_to_end
End-to-end: EGA computes directions and granular projection modifies weights.
tests/test_abliterate.py:2578
Methodtest_emit_records_durations
Verify _emit stores durations in _stage_durations dict.
tests/test_telemetry.py:579
Methodtest_empty
(self)
tests/test_metrics.py:50
Methodtest_empty_activations
(self)
tests/test_novel_analysis.py:202
Methodtest_empty_batch
(self)
tests/test_novel_analysis.py:472
Methodtest_empty_categories
(self)
tests/test_breakthrough_modules.py:378
Methodtest_empty_directions
Should handle empty input gracefully.
tests/test_logit_lens.py:140
Methodtest_empty_directions
(self)
tests/test_novel_analysis.py:281
Methodtest_empty_directions
(self)
tests/test_telemetry.py:246
Methodtest_empty_input
(self)
tests/test_abliterate.py:1666
Methodtest_empty_input
Should handle empty input gracefully.
tests/test_analysis.py:192
Methodtest_empty_input
Should handle empty input gracefully.
tests/test_analysis.py:340
Methodtest_empty_input
(self)
tests/test_breakthrough_modules.py:173
Methodtest_empty_input
(self)
tests/test_breakthrough_modules.py:280
Methodtest_empty_input
(self)
tests/test_breakthrough_modules.py:517
Methodtest_empty_input
(self)
tests/test_new_analysis_modules.py:624
Methodtest_empty_inputs
(self)
tests/test_novel_analysis.py:640
Methodtest_empty_list
(self)
tests/test_heretic_eval.py:89
Methodtest_empty_list
(self)
tests/test_refusal_detection.py:98
Methodtest_empty_list
(self)
tests/test_analysis_utils.py:13
Methodtest_empty_model_name
Empty string should fall through to Dense Standard.
tests/test_architecture_profiles.py:551
Methodtest_empty_overrides
(self)
tests/test_architecture_profiles.py:399
Methodtest_empty_pipeline
(self)
tests/test_defense_robustness.py:74
Methodtest_empty_responses
(self)
tests/test_refusal_detection.py:133
Methodtest_empty_responses
(self)
tests/test_advanced_metrics.py:73
Methodtest_empty_responses_not_refusals
Per Zou et al. logic, empty responses have no substring match = not refusals.
tests/test_heretic_eval.py:84
Methodtest_empty_string
(self)
tests/test_refusal_detection.py:23
Methodtest_empty_string_not_refusal
(self)
tests/test_advanced_metrics.py:76
Methodtest_enable_programmatically
(self)
tests/test_telemetry.py:75
Methodtest_enable_via_env_explicit
(self)
tests/test_telemetry.py:70
Methodtest_enabled_by_default_on_hf_spaces
(self)
tests/test_telemetry.py:51
Methodtest_energy_removed
Top rows should capture most of the refusal energy.
tests/test_novel_analysis.py:523
Methodtest_entangled_layers_skipped
(self)
tests/test_informed_pipeline.py:250
Methodtest_entanglement_gate
(self, pipeline)
tests/test_informed_pipeline.py:117
Methodtest_entanglement_map
(self)
tests/test_defense_robustness.py:135
Methodtest_entropy_nonnegative
Logit effect entropy should be non-negative.
tests/test_logit_lens.py:154
Methodtest_enumerate
(self, handle)
tests/test_strategies.py:77
Methodtest_enumerate
(self, handle)
tests/test_strategies.py:112
Methodtest_enumerate
(self, handle)
tests/test_strategies.py:141
Methodtest_enumerate
(self, handle)
tests/test_strategies.py:162
Methodtest_environment_info
(self)
tests/test_telemetry.py:145
Functiontest_estimate_total_params_uses_nested_text_config
()
tests/test_model_profile.py:11
Methodtest_estimated_passes
(self)
tests/test_breakthrough_modules.py:287
Methodtest_excise_advanced_norm_preserving
Advanced method with norm preservation should maintain weight norms.
tests/test_abliterate.py:1843
Methodtest_excise_basic
Basic method should modify weights.
tests/test_abliterate.py:1809
Methodtest_excise_details
(self)
tests/test_telemetry.py:162
Methodtest_expected_benchmarks
(self)
tests/test_heretic_eval.py:305
Methodtest_expert_input_projections_projected
Expert input projections (up_proj, gate_proj) should also be modified.
tests/test_abliterate.py:527
Methodtest_expert_transplant
Expert transplant should overwrite safety expert weights with capability average.
tests/test_abliterate.py:1552
Methodtest_explicit_layer_mapping
(self)
tests/test_breakthrough_modules.py:477
Methodtest_explicit_overrides_method
(self)
tests/test_abliterate.py:203
Methodtest_explicit_overrides_new_params
(self)
tests/test_abliterate_extended.py:127
Methodtest_extract_all_layers
Should extract directions for all provided layers.
tests/test_analysis.py:80
Methodtest_extract_all_layers
(self)
tests/test_new_analysis_modules.py:463
Methodtest_extract_number
(self)
tests/test_benchmarks.py:176
Methodtest_falcon3_is_not_reasoning
falcon3 must NOT match 'o3' reasoning pattern.
tests/test_architecture_profiles.py:167
Methodtest_fallback_to_next_key
(self)
tests/test_heretic_eval.py:357
Methodtest_fallback_to_shared_on_fused
Experts without per-expert direction should use shared direction.
tests/test_abliterate.py:2445
Methodtest_fallback_when_no_template
Should fall back to raw prompts when template is not configured.
tests/test_abliterate_extended.py:244
Methodtest_falls_back_to_shared_direction
Experts without per-expert direction should use shared direction.
tests/test_abliterate.py:2292
Methodtest_false_positive_sorry
I'm sorry to hear that' followed by helpful content.
tests/test_edge_cases.py:300
Methodtest_feature_clustering
(self)
tests/test_new_analysis_modules.py:391
Methodtest_feature_extraction
(self)
tests/test_novel_analysis.py:269
Methodtest_fewshot_settings_match_arditi
Few-shot settings should match Arditi et al. / Open LLM Leaderboard.
tests/test_heretic_eval.py:309
Methodtest_filename_format
(self, tmp_path)
tests/test_community.py:155
Methodtest_filters_unknown_config_keys
(self)
tests/test_telemetry.py:124
Methodtest_finds_gate
Should find a router named 'gate'.
tests/test_abliterate.py:1958
Methodtest_finds_router
Should find a router named 'router'.
tests/test_abliterate.py:1972
Methodtest_first_token_kl_different
First-token KL of different distributions should be positive.
tests/test_advanced_metrics.py:129
Methodtest_first_token_kl_identical
First-token KL of identical distributions should be 0.
tests/test_advanced_metrics.py:123
Methodtest_five_percent_threshold_includes_more
Layers between 5% and 10% of max should now be included.
tests/test_abliterate.py:434
Methodtest_five_standard_benchmarks
(self)
tests/test_heretic_eval.py:296
Methodtest_flat_manifold_detection
When activations are purely linear, curvature should be near zero.
tests/test_breakthrough_modules.py:189
Methodtest_format_analysis
(self)
tests/test_novel_analysis.py:619
Methodtest_format_cross_category
(self)
tests/test_causal_and_transfer.py:497
Methodtest_format_cross_model
(self)
tests/test_causal_and_transfer.py:488
Methodtest_format_decomposition
(self)
tests/test_causal_and_transfer.py:254
Methodtest_format_default
(self, insights)
tests/test_informed_pipeline.py:302
Methodtest_format_entanglement
(self)
tests/test_defense_robustness.py:163
Methodtest_format_imprint
(self)
tests/test_novel_analysis.py:307
← previousnext →1,201–1,300 of 1,841, ranked by callers