Commit Graph

  • e093b117a7 Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 14:27:15 +00:00
  • b70a3d89f1 Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 14:16:01 +00:00
  • 21812c7757 Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 14:12:59 +00:00
  • cac07f919e Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 14:12:29 +00:00
  • 8850354876 Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 14:10:25 +00:00
  • 3bd747856b Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 14:10:11 +00:00
  • 8a9b1e3fc6 Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 14:10:07 +00:00
  • dc78ccd5cf Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 14:08:51 +00:00
  • 84ba97198c Merge pull request #28 from lightbroker/model-support-expansion Adam Wilson 2025-08-19 07:50:50 -06:00
  • df9ae66a12 err handling Adam Wilson 2025-08-19 07:50:30 -06:00
  • 5cb4f063e3 Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 13:14:02 +00:00
  • 67239a27ea Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 12:56:29 +00:00
  • 2734ff45f9 Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 12:48:11 +00:00
  • 66c3233314 Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 12:45:12 +00:00
  • 563161d375 Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 12:43:33 +00:00
  • e9132128ea Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 12:40:52 +00:00
  • 3b204ecb60 Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 12:38:07 +00:00
  • 44d61f2007 Auto-generated files from microsoft/Phi-3-mini-4k-instruct range 1-20 batch 4 [skip ci] Adam Wilson 2025-08-19 12:36:40 +00:00
  • 623ad5aaa6 Merge pull request #27 from lightbroker/model-support-expansion Adam Wilson 2025-08-19 06:22:43 -06:00
  • 71bba11550 checkout new branch during workflow Adam Wilson 2025-08-19 06:22:14 -06:00
  • 60d092b0af Merge pull request #26 from lightbroker/model-support-expansion Adam Wilson 2025-08-18 21:30:44 -06:00
  • a238893375 os.makedirs() Adam Wilson 2025-08-18 21:30:15 -06:00
  • 368c9e68fb fix path Adam Wilson 2025-08-18 21:10:50 -06:00
  • 28a962facd support casing Adam Wilson 2025-08-18 20:48:43 -06:00
  • fad7035b32 batch test workflows for OpenELM Adam Wilson 2025-08-18 20:38:59 -06:00
  • 6de1a82575 fix llama Adam Wilson 2025-08-18 20:27:56 -06:00
  • 196148ffe0 Merge pull request #25 from lightbroker/model-support-expansion Adam Wilson 2025-08-18 20:25:03 -06:00
  • 42e4962063 base classes Adam Wilson 2025-08-18 20:24:36 -06:00
  • 42054a08dd Merge pull request #24 from lightbroker/model-support-expansion Adam Wilson 2025-08-18 20:05:52 -06:00
  • a636b7fbf7 fix test number file path; support smaller runs Adam Wilson 2025-08-18 20:05:24 -06:00
  • 51c0a92f6b Merge pull request #23 from lightbroker/model-support-expansion Adam Wilson 2025-08-18 19:47:10 -06:00
  • 2694962fc8 try smaller batches Adam Wilson 2025-08-18 19:46:51 -06:00
  • 7b466d9bf2 try smaller batches Adam Wilson 2025-08-18 19:46:11 -06:00
  • c142235923 Merge pull request #22 from lightbroker/model-support-expansion Adam Wilson 2025-08-18 18:00:45 -06:00
  • 8c6230e0dc try smaller batches Adam Wilson 2025-08-18 18:00:08 -06:00
  • 09eac1f050 try smaller batches Adam Wilson 2025-08-18 17:36:01 -06:00
  • 0411049d6b matrix strategy for tests; remove dead code Adam Wilson 2025-08-18 16:31:32 -06:00
  • 010933aa59 matrix strategy for tests Adam Wilson 2025-08-18 16:12:31 -06:00
  • b1a2672d14 test result folders Adam Wilson 2025-08-18 15:50:07 -06:00
  • a1c7313833 fix for template selection Adam Wilson 2025-08-18 15:16:27 -06:00
  • 2e1c43b819 unneeded param Adam Wilson 2025-08-18 14:49:05 -06:00
  • 73b25d2ae4 fix arg Adam Wilson 2025-08-18 14:38:37 -06:00
  • 3d335d047c dead code Adam Wilson 2025-08-18 14:31:46 -06:00
  • fb95e8afea path Adam Wilson 2025-08-18 14:04:06 -06:00
  • d7bffb85e1 env Adam Wilson 2025-08-18 13:55:31 -06:00
  • a48ea0c626 Merge pull request #21 from lightbroker/model-support-expansion Adam Wilson 2025-08-18 13:43:11 -06:00
  • 67620f780b test 4 - GitHub Actions Adam Wilson 2025-08-18 13:42:44 -06:00
  • 9f3b8b6b07 reorganize tests Adam Wilson 2025-08-18 13:28:01 -06:00
  • 36c11703cb dynamic template and model selection Adam Wilson 2025-08-18 11:22:50 -06:00
  • a1e07e6a4f support multiple templates Adam Wilson 2025-08-16 19:47:26 -06:00
  • e138a14d5f support multiple templates Adam Wilson 2025-08-16 19:30:05 -06:00
  • d5beccb774 remove additional reflexion support Adam Wilson 2025-08-16 19:26:12 -06:00
  • 7a510f03fa commit step Adam Wilson 2025-08-16 19:25:51 -06:00
  • e014e6c321 remove reflexion Adam Wilson 2025-08-16 19:25:34 -06:00
  • a40c655334 old reference Adam Wilson 2025-08-16 19:08:21 -06:00
  • b0ae16af13 Merge pull request #20 from lightbroker/model-support-expansion Adam Wilson 2025-08-16 18:59:39 -06:00
  • 1eadd81d77 new test for GH actions Adam Wilson 2025-08-16 18:57:08 -06:00
  • 82c987404b templates Adam Wilson 2025-08-16 16:00:35 -06:00
  • 11028c6b4e + model support (Apple OpenELM 270M Instruct, Meta TinyLlama 1.1B Chat) Adam Wilson 2025-08-16 12:33:18 -06:00
  • 935e5b9a5b updated test results table with metrics from Aug 5 re-run of test_4 Adam Wilson 2025-08-05 18:30:26 -06:00
  • 757eb64c35 new test run results for test_04_malicious_prompts_rag_and_cot Adam Wilson 2025-08-05 18:24:36 -06:00
  • ab7cd56d7e test archival Adam Wilson 2025-08-05 18:24:15 -06:00
  • 1f0bd51ff3 fix template Adam Wilson 2025-08-05 18:23:52 -06:00
  • b7c22809a6 archiving tests Adam Wilson 2025-08-05 18:23:06 -06:00
  • 4e161e850e template Adam Wilson 2025-08-05 15:09:39 -06:00
  • 75c048c00d :( Adam Wilson 2025-08-05 15:07:37 -06:00
  • 24832ef812 CoT + RAG template Adam Wilson 2025-08-05 14:44:02 -06:00
  • ae3dd68f12 violation rate threshold = 15% Adam Wilson 2025-08-04 22:04:10 -06:00
  • c94fbb3b9d <= Adam Wilson 2025-08-04 21:53:02 -06:00
  • a49fef62c5 use percentage not floating point Adam Wilson 2025-08-04 21:51:24 -06:00
  • 74f99a36ec unneeded "build" phase Adam Wilson 2025-08-04 21:37:35 -06:00
  • 2e69881672 Merge pull request #18 from lightbroker/development Adam Wilson 2025-08-04 21:24:29 -06:00
  • 4c317634a3 fix order; add PR trigger Adam Wilson 2025-08-04 21:23:56 -06:00
  • 14c4a0adc0 Merge pull request #17 from lightbroker/development Adam Wilson 2025-08-04 08:38:10 -06:00
  • 865a4b923a job dependencies Adam Wilson 2025-08-04 08:02:59 -06:00
  • 4bd7c47699 Merge pull request #16 from lightbroker/development Adam Wilson 2025-08-04 07:59:56 -06:00
  • 3b5c7b9f69 CI/CD test Adam Wilson 2025-08-04 07:59:02 -06:00
  • 5252342701 add violation rate column to results table, based on 100 - (< thresh %) Adam Wilson 2025-08-02 10:18:13 -06:00
  • b65e531049 results table Adam Wilson 2025-08-01 20:00:14 -06:00
  • 6d5d53bad5 summary of test results (tests 0-4) Adam Wilson 2025-08-01 19:38:19 -06:00
  • bef6ba9ef5 JSON results: test_00_benign_prompts_no_mitigation Adam Wilson 2025-07-30 12:19:53 -06:00
  • 0171af7c94 fix confusing log message Adam Wilson 2025-07-30 11:16:24 -06:00
  • 24c6eaae28 JSON schema script Adam Wilson 2025-07-30 11:13:20 -06:00
  • 378aea7a66 100 math prompts, not 150 Adam Wilson 2025-07-30 11:13:09 -06:00
  • 1624019272 JSON results: test_01_malicious_prompts_no_mitigation Adam Wilson 2025-07-30 11:08:46 -06:00
  • 28518e6c17 fixes for basic prompt template Adam Wilson 2025-07-30 07:17:48 -06:00
  • 0aaa4036df JSON results: test_02_malicious_prompts_cot Adam Wilson 2025-07-30 05:53:33 -06:00
  • c7deca79da JSON results: test_03_malicious_prompts_rag Adam Wilson 2025-07-29 22:13:56 -06:00
  • 4f26df0755 JSON results: test_04_malicious_prompts_rag_and_cot Adam Wilson 2025-07-29 15:27:39 -06:00
  • cf4e2d48fa JSON results: test_05_malicious_prompts_cot_with_reflexion Adam Wilson 2025-07-29 11:54:00 -06:00
  • bf8396bb03 JSON results: test_06_malicious_prompts_rag_with_reflexion Adam Wilson 2025-07-29 06:09:29 -06:00
  • df6d5e644b JSON results: test_07_malicious_prompts_rag_and_cot_reflexion Adam Wilson 2025-07-28 20:51:31 -06:00
  • 780a4e529f archive tests Adam Wilson 2025-07-28 15:27:04 -06:00
  • 51e966e1ec escape braces Adam Wilson 2025-07-28 14:55:32 -06:00
  • 047c161006 archive test results Adam Wilson 2025-07-28 13:43:28 -06:00
  • 058b281f25 default risk threshold; scores lower down in JSON serialization Adam Wilson 2025-07-28 13:40:54 -06:00
  • d39d7f1f1f don't send back alternate after guidelines Adam Wilson 2025-07-28 13:29:21 -06:00
  • 393b76faf4 small fix Adam Wilson 2025-07-28 13:19:50 -06:00
  • bc119ed50e don't overprocess results Adam Wilson 2025-07-28 13:10:03 -06:00
  • 24e28c0182 fix JSON serialization for logging test result Adam Wilson 2025-07-28 12:06:56 -06:00