inspect-ai 0.3.62__py3-none-any.whl → 0.3.64__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (518) hide show
  1. inspect_ai/_cli/cache.py +8 -7
  2. inspect_ai/_cli/common.py +0 -12
  3. inspect_ai/_cli/eval.py +32 -4
  4. inspect_ai/_cli/info.py +1 -0
  5. inspect_ai/_cli/list.py +1 -1
  6. inspect_ai/_cli/log.py +2 -0
  7. inspect_ai/_cli/main.py +1 -1
  8. inspect_ai/_cli/sandbox.py +4 -1
  9. inspect_ai/_cli/score.py +181 -32
  10. inspect_ai/_cli/trace.py +10 -0
  11. inspect_ai/_cli/view.py +4 -2
  12. inspect_ai/_display/core/active.py +2 -3
  13. inspect_ai/_display/core/config.py +7 -1
  14. inspect_ai/_display/textual/widgets/samples.py +4 -3
  15. inspect_ai/_display/textual/widgets/sandbox.py +6 -0
  16. inspect_ai/_eval/eval.py +104 -101
  17. inspect_ai/_eval/evalset.py +75 -75
  18. inspect_ai/_eval/loader.py +122 -12
  19. inspect_ai/_eval/registry.py +1 -1
  20. inspect_ai/_eval/run.py +14 -0
  21. inspect_ai/_eval/score.py +125 -36
  22. inspect_ai/_eval/task/log.py +105 -4
  23. inspect_ai/_eval/task/results.py +92 -38
  24. inspect_ai/_eval/task/run.py +9 -2
  25. inspect_ai/_eval/task/sandbox.py +35 -2
  26. inspect_ai/_eval/task/task.py +49 -46
  27. inspect_ai/_util/constants.py +1 -1
  28. inspect_ai/_util/content.py +8 -0
  29. inspect_ai/_util/error.py +2 -0
  30. inspect_ai/_util/file.py +15 -1
  31. inspect_ai/_util/hash.py +1 -1
  32. inspect_ai/_util/logger.py +4 -2
  33. inspect_ai/_util/registry.py +7 -1
  34. inspect_ai/_view/view.py +1 -2
  35. inspect_ai/_view/www/.vscode/extensions.json +3 -0
  36. inspect_ai/_view/www/.vscode/settings.json +8 -0
  37. inspect_ai/_view/www/App.css +97 -29
  38. inspect_ai/_view/www/README.md +1 -1
  39. inspect_ai/_view/www/dist/assets/index.css +16663 -14674
  40. inspect_ai/_view/www/dist/assets/index.js +58808 -51348
  41. inspect_ai/_view/www/dist/index.html +1 -1
  42. inspect_ai/_view/www/index.html +2 -2
  43. inspect_ai/_view/www/log-schema.json +87 -73
  44. inspect_ai/_view/www/package.json +22 -4
  45. inspect_ai/_view/www/postcss.config.cjs +8 -9
  46. inspect_ai/_view/www/src/{App.mjs → App.tsx} +356 -365
  47. inspect_ai/_view/www/src/AppErrorBoundary.tsx +47 -0
  48. inspect_ai/_view/www/src/api/api-browser.ts +2 -2
  49. inspect_ai/_view/www/src/api/api-http.ts +3 -5
  50. inspect_ai/_view/www/src/api/api-vscode.ts +6 -6
  51. inspect_ai/_view/www/src/api/client-api.ts +4 -4
  52. inspect_ai/_view/www/src/api/index.ts +4 -4
  53. inspect_ai/_view/www/src/api/{Types.ts → types.ts} +25 -9
  54. inspect_ai/_view/www/src/appearance/colors.ts +9 -0
  55. inspect_ai/_view/www/src/appearance/fonts.ts +39 -0
  56. inspect_ai/_view/www/src/appearance/icons.ts +100 -0
  57. inspect_ai/_view/www/src/appearance/{Styles.mjs → styles.ts} +2 -32
  58. inspect_ai/_view/www/src/components/AnsiDisplay.tsx +198 -0
  59. inspect_ai/_view/www/src/components/AsciinemaPlayer.tsx +86 -0
  60. inspect_ai/_view/www/src/components/Card.css +60 -0
  61. inspect_ai/_view/www/src/components/Card.tsx +109 -0
  62. inspect_ai/_view/www/src/components/CopyButton.module.css +11 -0
  63. inspect_ai/_view/www/src/components/CopyButton.tsx +58 -0
  64. inspect_ai/_view/www/src/components/DownloadButton.css +4 -0
  65. inspect_ai/_view/www/src/components/DownloadButton.tsx +25 -0
  66. inspect_ai/_view/www/src/components/DownloadPanel.css +10 -0
  67. inspect_ai/_view/www/src/components/DownloadPanel.tsx +30 -0
  68. inspect_ai/_view/www/src/components/EmptyPanel.css +12 -0
  69. inspect_ai/_view/www/src/components/EmptyPanel.tsx +15 -0
  70. inspect_ai/_view/www/src/components/ErrorPanel.css +37 -0
  71. inspect_ai/_view/www/src/components/ErrorPanel.tsx +39 -0
  72. inspect_ai/_view/www/src/components/ExpandablePanel.css +40 -0
  73. inspect_ai/_view/www/src/components/ExpandablePanel.tsx +115 -0
  74. inspect_ai/_view/www/src/components/FindBand.css +49 -0
  75. inspect_ai/_view/www/src/components/FindBand.tsx +130 -0
  76. inspect_ai/_view/www/src/components/HumanBaselineView.css +41 -0
  77. inspect_ai/_view/www/src/components/HumanBaselineView.tsx +162 -0
  78. inspect_ai/_view/www/src/components/JsonPanel.css +20 -0
  79. inspect_ai/_view/www/src/components/JsonPanel.tsx +82 -0
  80. inspect_ai/_view/www/src/components/LabeledValue.css +20 -0
  81. inspect_ai/_view/www/src/components/LabeledValue.tsx +41 -0
  82. inspect_ai/_view/www/src/components/LargeModal.module.css +54 -0
  83. inspect_ai/_view/www/src/components/LargeModal.tsx +189 -0
  84. inspect_ai/_view/www/src/components/LightboxCarousel.css +95 -0
  85. inspect_ai/_view/www/src/components/LightboxCarousel.tsx +132 -0
  86. inspect_ai/_view/www/src/components/MarkdownDiv.css +3 -0
  87. inspect_ai/_view/www/src/components/MarkdownDiv.tsx +133 -0
  88. inspect_ai/_view/www/src/components/MessageBand.css +43 -0
  89. inspect_ai/_view/www/src/components/MessageBand.tsx +39 -0
  90. inspect_ai/_view/www/src/components/MorePopOver.css +0 -0
  91. inspect_ai/_view/www/src/components/MorePopOver.tsx +67 -0
  92. inspect_ai/_view/www/src/components/NavPills.module.css +18 -0
  93. inspect_ai/_view/www/src/components/NavPills.tsx +101 -0
  94. inspect_ai/_view/www/src/components/ProgressBar.module.css +37 -0
  95. inspect_ai/_view/www/src/components/ProgressBar.tsx +22 -0
  96. inspect_ai/_view/www/src/components/TabSet.module.css +40 -0
  97. inspect_ai/_view/www/src/components/TabSet.tsx +215 -0
  98. inspect_ai/_view/www/src/components/ToolButton.css +3 -0
  99. inspect_ai/_view/www/src/components/ToolButton.tsx +27 -0
  100. inspect_ai/_view/www/src/components/VirtualList.module.css +19 -0
  101. inspect_ai/_view/www/src/components/VirtualList.tsx +292 -0
  102. inspect_ai/_view/www/src/{index.js → index.tsx} +45 -19
  103. inspect_ai/_view/www/src/{log → logfile}/remoteLogFile.ts +3 -8
  104. inspect_ai/_view/www/src/{utils/remoteZipFile.mjs → logfile/remoteZipFile.ts} +86 -80
  105. inspect_ai/_view/www/src/metadata/MetaDataGrid.tsx +83 -0
  106. inspect_ai/_view/www/src/metadata/MetaDataView.module.css +35 -0
  107. inspect_ai/_view/www/src/metadata/MetaDataView.tsx +95 -0
  108. inspect_ai/_view/www/src/metadata/MetadataGrid.module.css +15 -0
  109. inspect_ai/_view/www/src/metadata/RenderedContent.module.css +12 -0
  110. inspect_ai/_view/www/src/{components/RenderedContent/RenderedContent.mjs → metadata/RenderedContent.tsx} +92 -73
  111. inspect_ai/_view/www/src/metadata/types.ts +18 -0
  112. inspect_ai/_view/www/src/plan/DatasetDetailView.module.css +3 -0
  113. inspect_ai/_view/www/src/plan/DatasetDetailView.tsx +37 -0
  114. inspect_ai/_view/www/src/plan/DetailStep.module.css +9 -0
  115. inspect_ai/_view/www/src/plan/DetailStep.tsx +31 -0
  116. inspect_ai/_view/www/src/plan/PlanCard.tsx +28 -0
  117. inspect_ai/_view/www/src/plan/PlanDetailView.module.css +48 -0
  118. inspect_ai/_view/www/src/plan/PlanDetailView.tsx +324 -0
  119. inspect_ai/_view/www/src/plan/ScorerDetailView.module.css +3 -0
  120. inspect_ai/_view/www/src/plan/ScorerDetailView.tsx +30 -0
  121. inspect_ai/_view/www/src/plan/SolverDetailView.module.css +15 -0
  122. inspect_ai/_view/www/src/plan/SolverDetailView.tsx +32 -0
  123. inspect_ai/_view/www/src/samples/InlineSampleDisplay.module.css +8 -0
  124. inspect_ai/_view/www/src/samples/InlineSampleDisplay.tsx +53 -0
  125. inspect_ai/_view/www/src/samples/SampleDialog.tsx +122 -0
  126. inspect_ai/_view/www/src/samples/SampleDisplay.module.css +29 -0
  127. inspect_ai/_view/www/src/samples/SampleDisplay.tsx +331 -0
  128. inspect_ai/_view/www/src/samples/SampleSummaryView.module.css +24 -0
  129. inspect_ai/_view/www/src/samples/SampleSummaryView.tsx +177 -0
  130. inspect_ai/_view/www/src/samples/SamplesTools.tsx +52 -0
  131. inspect_ai/_view/www/src/samples/chat/ChatMessage.module.css +29 -0
  132. inspect_ai/_view/www/src/samples/chat/ChatMessage.tsx +76 -0
  133. inspect_ai/_view/www/src/samples/chat/ChatMessageRenderer.tsx +60 -0
  134. inspect_ai/_view/www/src/samples/chat/ChatMessageRow.module.css +9 -0
  135. inspect_ai/_view/www/src/samples/chat/ChatMessageRow.tsx +57 -0
  136. inspect_ai/_view/www/src/samples/chat/ChatView.tsx +47 -0
  137. inspect_ai/_view/www/src/samples/chat/ChatViewVirtualList.module.css +4 -0
  138. inspect_ai/_view/www/src/samples/chat/ChatViewVirtualList.tsx +58 -0
  139. inspect_ai/_view/www/src/samples/chat/MessageContent.module.css +4 -0
  140. inspect_ai/_view/www/src/samples/chat/MessageContent.tsx +157 -0
  141. inspect_ai/_view/www/src/samples/chat/MessageContents.module.css +3 -0
  142. inspect_ai/_view/www/src/samples/chat/MessageContents.tsx +133 -0
  143. inspect_ai/_view/www/src/samples/chat/messages.ts +112 -0
  144. inspect_ai/_view/www/src/samples/chat/tools/ToolCallView.tsx +147 -0
  145. inspect_ai/_view/www/src/samples/chat/tools/ToolInput.module.css +14 -0
  146. inspect_ai/_view/www/src/samples/chat/tools/ToolInput.tsx +76 -0
  147. inspect_ai/_view/www/src/samples/chat/tools/ToolOutput.module.css +19 -0
  148. inspect_ai/_view/www/src/samples/chat/tools/ToolOutput.tsx +60 -0
  149. inspect_ai/_view/www/src/samples/chat/tools/ToolTitle.module.css +4 -0
  150. inspect_ai/_view/www/src/samples/chat/tools/ToolTitle.tsx +18 -0
  151. inspect_ai/_view/www/src/samples/chat/tools/tool.ts +92 -0
  152. inspect_ai/_view/www/src/samples/descriptor/samplesDescriptor.tsx +365 -0
  153. inspect_ai/_view/www/src/samples/descriptor/score/BooleanScoreDescriptor.module.css +22 -0
  154. inspect_ai/_view/www/src/samples/descriptor/score/BooleanScoreDescriptor.tsx +26 -0
  155. inspect_ai/_view/www/src/samples/descriptor/score/CategoricalScoreDescriptor.tsx +18 -0
  156. inspect_ai/_view/www/src/samples/descriptor/score/NumericScoreDescriptor.tsx +27 -0
  157. inspect_ai/_view/www/src/samples/descriptor/score/ObjectScoreDescriptor.module.css +18 -0
  158. inspect_ai/_view/www/src/samples/descriptor/score/ObjectScoreDescriptor.tsx +71 -0
  159. inspect_ai/_view/www/src/samples/descriptor/score/OtherScoreDescriptor.tsx +20 -0
  160. inspect_ai/_view/www/src/samples/descriptor/score/PassFailScoreDescriptor.module.css +28 -0
  161. inspect_ai/_view/www/src/samples/descriptor/score/PassFailScoreDescriptor.tsx +81 -0
  162. inspect_ai/_view/www/src/samples/descriptor/score/ScoreDescriptor.tsx +99 -0
  163. inspect_ai/_view/www/src/samples/descriptor/types.ts +55 -0
  164. inspect_ai/_view/www/src/samples/error/FlatSampleErrorView.module.css +19 -0
  165. inspect_ai/_view/www/src/samples/error/FlatSampleErrorView.tsx +22 -0
  166. inspect_ai/_view/www/src/samples/error/SampleErrorView.module.css +17 -0
  167. inspect_ai/_view/www/src/samples/error/SampleErrorView.tsx +31 -0
  168. inspect_ai/_view/www/src/samples/error/error.ts +15 -0
  169. inspect_ai/_view/www/src/samples/list/SampleFooter.module.css +9 -0
  170. inspect_ai/_view/www/src/samples/list/SampleFooter.tsx +14 -0
  171. inspect_ai/_view/www/src/samples/list/SampleHeader.module.css +13 -0
  172. inspect_ai/_view/www/src/samples/list/SampleHeader.tsx +36 -0
  173. inspect_ai/_view/www/src/samples/list/SampleList.module.css +11 -0
  174. inspect_ai/_view/www/src/samples/list/SampleList.tsx +247 -0
  175. inspect_ai/_view/www/src/samples/list/SampleRow.module.css +33 -0
  176. inspect_ai/_view/www/src/samples/list/SampleRow.tsx +98 -0
  177. inspect_ai/_view/www/src/samples/list/SampleSeparator.module.css +6 -0
  178. inspect_ai/_view/www/src/samples/list/SampleSeparator.tsx +24 -0
  179. inspect_ai/_view/www/src/samples/sample-tools/EpochFilter.module.css +9 -0
  180. inspect_ai/_view/www/src/samples/sample-tools/EpochFilter.tsx +51 -0
  181. inspect_ai/_view/www/src/samples/sample-tools/SelectScorer.module.css +16 -0
  182. inspect_ai/_view/www/src/samples/sample-tools/SelectScorer.tsx +175 -0
  183. inspect_ai/_view/www/src/samples/sample-tools/SortFilter.module.css +9 -0
  184. inspect_ai/_view/www/src/samples/sample-tools/SortFilter.tsx +186 -0
  185. inspect_ai/_view/www/src/samples/{tools/filters.mjs → sample-tools/filters.ts} +86 -81
  186. inspect_ai/_view/www/src/samples/sample-tools/sample-filter/SampleFilter.module.css +16 -0
  187. inspect_ai/_view/www/src/samples/sample-tools/sample-filter/SampleFilter.tsx +288 -0
  188. inspect_ai/_view/www/src/samples/sample-tools/sample-filter/completions.ts +346 -0
  189. inspect_ai/_view/www/src/samples/sample-tools/sample-filter/language.ts +19 -0
  190. inspect_ai/_view/www/src/samples/sample-tools/sample-filter/tokenize.ts +97 -0
  191. inspect_ai/_view/www/src/samples/{SampleLimit.mjs → sampleLimit.ts} +3 -6
  192. inspect_ai/_view/www/src/samples/scores/SampleScoreView.module.css +53 -0
  193. inspect_ai/_view/www/src/samples/scores/SampleScoreView.tsx +168 -0
  194. inspect_ai/_view/www/src/samples/scores/SampleScores.module.css +5 -0
  195. inspect_ai/_view/www/src/samples/scores/SampleScores.tsx +37 -0
  196. inspect_ai/_view/www/src/samples/transcript/ApprovalEventView.tsx +66 -0
  197. inspect_ai/_view/www/src/samples/transcript/ErrorEventView.tsx +51 -0
  198. inspect_ai/_view/www/src/samples/transcript/InfoEventView.module.css +3 -0
  199. inspect_ai/_view/www/src/samples/transcript/InfoEventView.tsx +54 -0
  200. inspect_ai/_view/www/src/samples/transcript/InputEventView.tsx +48 -0
  201. inspect_ai/_view/www/src/samples/transcript/LoggerEventView.module.css +6 -0
  202. inspect_ai/_view/www/src/samples/transcript/LoggerEventView.tsx +36 -0
  203. inspect_ai/_view/www/src/samples/transcript/ModelEventView.module.css +43 -0
  204. inspect_ai/_view/www/src/samples/transcript/ModelEventView.tsx +223 -0
  205. inspect_ai/_view/www/src/samples/transcript/SampleInitEventView.module.css +23 -0
  206. inspect_ai/_view/www/src/samples/transcript/SampleInitEventView.tsx +112 -0
  207. inspect_ai/_view/www/src/samples/transcript/SampleLimitEventView.tsx +75 -0
  208. inspect_ai/_view/www/src/samples/transcript/SampleTranscript.tsx +22 -0
  209. inspect_ai/_view/www/src/samples/transcript/ScoreEventView.module.css +15 -0
  210. inspect_ai/_view/www/src/samples/transcript/ScoreEventView.tsx +100 -0
  211. inspect_ai/_view/www/src/samples/transcript/StepEventView.tsx +171 -0
  212. inspect_ai/_view/www/src/samples/transcript/SubtaskEventView.module.css +19 -0
  213. inspect_ai/_view/www/src/samples/transcript/SubtaskEventView.tsx +133 -0
  214. inspect_ai/_view/www/src/samples/transcript/ToolEventView.module.css +10 -0
  215. inspect_ai/_view/www/src/samples/transcript/ToolEventView.tsx +92 -0
  216. inspect_ai/_view/www/src/samples/transcript/TranscriptView.module.css +49 -0
  217. inspect_ai/_view/www/src/samples/transcript/TranscriptView.tsx +449 -0
  218. inspect_ai/_view/www/src/samples/transcript/event/EventNav.module.css +5 -0
  219. inspect_ai/_view/www/src/samples/transcript/event/EventNav.tsx +43 -0
  220. inspect_ai/_view/www/src/samples/transcript/event/EventNavs.module.css +3 -0
  221. inspect_ai/_view/www/src/samples/transcript/event/EventNavs.tsx +39 -0
  222. inspect_ai/_view/www/src/samples/transcript/event/EventPanel.module.css +25 -0
  223. inspect_ai/_view/www/src/samples/transcript/event/EventPanel.tsx +191 -0
  224. inspect_ai/_view/www/src/samples/transcript/event/EventRow.module.css +13 -0
  225. inspect_ai/_view/www/src/samples/transcript/event/EventRow.tsx +32 -0
  226. inspect_ai/_view/www/src/samples/transcript/event/EventSection.module.css +8 -0
  227. inspect_ai/_view/www/src/samples/transcript/event/EventSection.tsx +29 -0
  228. inspect_ai/_view/www/src/samples/transcript/state/StateDiffView.tsx +67 -0
  229. inspect_ai/_view/www/src/samples/transcript/state/StateEventRenderers.tsx +285 -0
  230. inspect_ai/_view/www/src/samples/transcript/state/StateEventRenders.module.css +10 -0
  231. inspect_ai/_view/www/src/samples/transcript/state/StateEventView.module.css +9 -0
  232. inspect_ai/_view/www/src/samples/transcript/state/StateEventView.tsx +346 -0
  233. inspect_ai/_view/www/src/samples/transcript/types.ts +58 -0
  234. inspect_ai/_view/www/src/types/log.d.ts +108 -19
  235. inspect_ai/_view/www/src/types/prism.d.ts +11 -0
  236. inspect_ai/_view/www/src/types.ts +71 -0
  237. inspect_ai/_view/www/src/usage/ModelTokenTable.tsx +28 -0
  238. inspect_ai/_view/www/src/usage/ModelUsagePanel.module.css +24 -0
  239. inspect_ai/_view/www/src/usage/ModelUsagePanel.tsx +97 -0
  240. inspect_ai/_view/www/src/usage/TokenTable.module.css +17 -0
  241. inspect_ai/_view/www/src/usage/TokenTable.tsx +91 -0
  242. inspect_ai/_view/www/src/usage/UsageCard.module.css +15 -0
  243. inspect_ai/_view/www/src/usage/UsageCard.tsx +67 -0
  244. inspect_ai/_view/www/src/utils/attachments.ts +42 -0
  245. inspect_ai/_view/www/src/utils/{Base64.mjs → base64.ts} +1 -6
  246. inspect_ai/_view/www/src/{components/Browser.mjs → utils/browser.ts} +0 -1
  247. inspect_ai/_view/www/src/utils/debugging.ts +28 -0
  248. inspect_ai/_view/www/src/utils/dom.ts +30 -0
  249. inspect_ai/_view/www/src/utils/format.ts +194 -0
  250. inspect_ai/_view/www/src/utils/git.ts +7 -0
  251. inspect_ai/_view/www/src/utils/html.ts +6 -0
  252. inspect_ai/_view/www/src/utils/http.ts +14 -0
  253. inspect_ai/_view/www/src/utils/{Path.mjs → path.ts} +2 -9
  254. inspect_ai/_view/www/src/utils/{Print.mjs → print.ts} +34 -26
  255. inspect_ai/_view/www/src/utils/queue.ts +51 -0
  256. inspect_ai/_view/www/src/utils/sync.ts +114 -0
  257. inspect_ai/_view/www/src/utils/{Type.mjs → type.ts} +3 -6
  258. inspect_ai/_view/www/src/utils/vscode.ts +13 -0
  259. inspect_ai/_view/www/src/workspace/WorkSpace.tsx +324 -0
  260. inspect_ai/_view/www/src/workspace/WorkSpaceView.module.css +33 -0
  261. inspect_ai/_view/www/src/workspace/WorkSpaceView.tsx +158 -0
  262. inspect_ai/_view/www/src/workspace/error/TaskErrorPanel.module.css +3 -0
  263. inspect_ai/_view/www/src/workspace/error/TaskErrorPanel.tsx +28 -0
  264. inspect_ai/_view/www/src/workspace/navbar/Navbar.module.css +54 -0
  265. inspect_ai/_view/www/src/workspace/navbar/Navbar.tsx +68 -0
  266. inspect_ai/_view/www/src/workspace/navbar/PrimaryBar.module.css +52 -0
  267. inspect_ai/_view/www/src/workspace/navbar/PrimaryBar.tsx +114 -0
  268. inspect_ai/_view/www/src/workspace/navbar/ResultsPanel.module.css +90 -0
  269. inspect_ai/_view/www/src/workspace/navbar/ResultsPanel.tsx +180 -0
  270. inspect_ai/_view/www/src/workspace/navbar/SecondaryBar.module.css +28 -0
  271. inspect_ai/_view/www/src/workspace/navbar/SecondaryBar.tsx +226 -0
  272. inspect_ai/_view/www/src/workspace/navbar/StatusPanel.module.css +14 -0
  273. inspect_ai/_view/www/src/workspace/navbar/StatusPanel.tsx +61 -0
  274. inspect_ai/_view/www/src/workspace/sidebar/EvalStatus.module.css +15 -0
  275. inspect_ai/_view/www/src/workspace/sidebar/EvalStatus.tsx +71 -0
  276. inspect_ai/_view/www/src/workspace/sidebar/LogDirectoryTitleView.module.css +5 -0
  277. inspect_ai/_view/www/src/workspace/sidebar/LogDirectoryTitleView.tsx +56 -0
  278. inspect_ai/_view/www/src/workspace/sidebar/Sidebar.module.css +68 -0
  279. inspect_ai/_view/www/src/workspace/sidebar/Sidebar.tsx +85 -0
  280. inspect_ai/_view/www/src/workspace/sidebar/SidebarLogEntry.module.css +29 -0
  281. inspect_ai/_view/www/src/workspace/sidebar/SidebarLogEntry.tsx +95 -0
  282. inspect_ai/_view/www/src/workspace/sidebar/SidebarScoreView.module.css +23 -0
  283. inspect_ai/_view/www/src/workspace/sidebar/SidebarScoreView.tsx +43 -0
  284. inspect_ai/_view/www/src/workspace/sidebar/SidebarScoresView.module.css +35 -0
  285. inspect_ai/_view/www/src/workspace/sidebar/SidebarScoresView.tsx +63 -0
  286. inspect_ai/_view/www/src/workspace/tabs/InfoTab.module.css +0 -0
  287. inspect_ai/_view/www/src/workspace/tabs/InfoTab.tsx +70 -0
  288. inspect_ai/_view/www/src/workspace/tabs/JsonTab.module.css +5 -0
  289. inspect_ai/_view/www/src/workspace/tabs/JsonTab.tsx +46 -0
  290. inspect_ai/_view/www/src/workspace/tabs/SamplesTab.tsx +204 -0
  291. inspect_ai/_view/www/src/workspace/tabs/grouping.ts +195 -0
  292. inspect_ai/_view/www/src/workspace/tabs/types.ts +19 -0
  293. inspect_ai/_view/www/src/workspace/types.ts +10 -0
  294. inspect_ai/_view/www/src/workspace/utils.ts +34 -0
  295. inspect_ai/_view/www/tsconfig.json +23 -9
  296. inspect_ai/_view/www/vite.config.js +8 -17
  297. inspect_ai/_view/www/yarn.lock +627 -556
  298. inspect_ai/approval/_approval.py +2 -0
  299. inspect_ai/approval/_approver.py +4 -4
  300. inspect_ai/approval/_auto.py +1 -1
  301. inspect_ai/approval/_human/approver.py +3 -0
  302. inspect_ai/approval/_policy.py +5 -0
  303. inspect_ai/approval/_registry.py +2 -2
  304. inspect_ai/dataset/_dataset.py +64 -37
  305. inspect_ai/dataset/_sources/__init__.py +0 -0
  306. inspect_ai/dataset/_sources/csv.py +20 -12
  307. inspect_ai/dataset/_sources/file.py +4 -0
  308. inspect_ai/dataset/_sources/hf.py +39 -29
  309. inspect_ai/dataset/_sources/json.py +17 -9
  310. inspect_ai/log/__init__.py +2 -0
  311. inspect_ai/log/_convert.py +3 -3
  312. inspect_ai/log/_file.py +24 -9
  313. inspect_ai/log/_log.py +101 -13
  314. inspect_ai/log/_message.py +4 -2
  315. inspect_ai/log/_recorders/file.py +4 -0
  316. inspect_ai/log/_recorders/json.py +5 -7
  317. inspect_ai/log/_recorders/recorder.py +3 -0
  318. inspect_ai/log/_transcript.py +19 -8
  319. inspect_ai/model/__init__.py +2 -0
  320. inspect_ai/model/_cache.py +39 -21
  321. inspect_ai/model/_call_tools.py +4 -3
  322. inspect_ai/model/_chat_message.py +14 -4
  323. inspect_ai/model/_generate_config.py +1 -1
  324. inspect_ai/model/_model.py +31 -24
  325. inspect_ai/model/_model_output.py +14 -1
  326. inspect_ai/model/_openai.py +10 -18
  327. inspect_ai/model/_providers/anthropic.py +3 -3
  328. inspect_ai/model/_providers/google.py +9 -5
  329. inspect_ai/model/_providers/openai.py +5 -9
  330. inspect_ai/model/_providers/openai_o1.py +3 -5
  331. inspect_ai/model/_providers/openrouter.py +86 -0
  332. inspect_ai/model/_providers/providers.py +11 -0
  333. inspect_ai/scorer/__init__.py +6 -1
  334. inspect_ai/scorer/_answer.py +7 -7
  335. inspect_ai/scorer/_classification.py +38 -18
  336. inspect_ai/scorer/_common.py +2 -8
  337. inspect_ai/scorer/_match.py +4 -5
  338. inspect_ai/scorer/_metric.py +87 -28
  339. inspect_ai/scorer/_metrics/__init__.py +3 -3
  340. inspect_ai/scorer/_metrics/accuracy.py +8 -10
  341. inspect_ai/scorer/_metrics/mean.py +3 -17
  342. inspect_ai/scorer/_metrics/std.py +111 -30
  343. inspect_ai/scorer/_model.py +12 -12
  344. inspect_ai/scorer/_pattern.py +3 -3
  345. inspect_ai/scorer/_reducer/reducer.py +36 -21
  346. inspect_ai/scorer/_reducer/registry.py +2 -2
  347. inspect_ai/scorer/_reducer/types.py +7 -1
  348. inspect_ai/scorer/_score.py +11 -1
  349. inspect_ai/scorer/_scorer.py +110 -16
  350. inspect_ai/solver/__init__.py +1 -1
  351. inspect_ai/solver/_basic_agent.py +19 -22
  352. inspect_ai/solver/_bridge/__init__.py +0 -3
  353. inspect_ai/solver/_bridge/bridge.py +3 -3
  354. inspect_ai/solver/_chain.py +1 -2
  355. inspect_ai/solver/_critique.py +3 -3
  356. inspect_ai/solver/_fork.py +2 -2
  357. inspect_ai/solver/_human_agent/__init__.py +0 -0
  358. inspect_ai/solver/_human_agent/agent.py +5 -8
  359. inspect_ai/solver/_human_agent/commands/clock.py +14 -10
  360. inspect_ai/solver/_human_agent/commands/note.py +1 -1
  361. inspect_ai/solver/_human_agent/commands/score.py +0 -11
  362. inspect_ai/solver/_multiple_choice.py +38 -26
  363. inspect_ai/solver/_prompt.py +7 -7
  364. inspect_ai/solver/_solver.py +53 -52
  365. inspect_ai/solver/_task_state.py +80 -69
  366. inspect_ai/solver/_use_tools.py +9 -9
  367. inspect_ai/tool/__init__.py +4 -1
  368. inspect_ai/tool/_tool.py +43 -14
  369. inspect_ai/tool/_tool_call.py +6 -2
  370. inspect_ai/tool/_tool_choice.py +3 -1
  371. inspect_ai/tool/_tool_def.py +10 -8
  372. inspect_ai/tool/_tool_params.py +24 -0
  373. inspect_ai/tool/_tool_with.py +7 -7
  374. inspect_ai/tool/_tools/__init__.py +0 -0
  375. inspect_ai/tool/{beta → _tools}/_computer/_common.py +2 -2
  376. inspect_ai/tool/{beta → _tools}/_computer/_computer.py +13 -5
  377. inspect_ai/tool/_tools/_computer/_resources/tool/__init__.py +0 -0
  378. inspect_ai/tool/{beta → _tools}/_computer/_resources/tool/_x11_client.py +1 -1
  379. inspect_ai/tool/_tools/_computer/_resources/tool/requirements.txt +0 -0
  380. inspect_ai/tool/_tools/_execute.py +23 -11
  381. inspect_ai/tool/_tools/_web_browser/_resources/README.md +2 -2
  382. inspect_ai/tool/_tools/_web_browser/_web_browser.py +5 -3
  383. inspect_ai/tool/_tools/_web_search.py +7 -5
  384. inspect_ai/tool/beta.py +3 -0
  385. inspect_ai/util/_concurrency.py +3 -3
  386. inspect_ai/util/_panel.py +2 -0
  387. inspect_ai/util/_resource.py +12 -12
  388. inspect_ai/util/_sandbox/docker/compose.py +23 -20
  389. inspect_ai/util/_sandbox/docker/config.py +2 -1
  390. inspect_ai/util/_sandbox/docker/docker.py +42 -86
  391. inspect_ai/util/_sandbox/docker/service.py +100 -0
  392. inspect_ai/util/_sandbox/environment.py +99 -96
  393. inspect_ai/util/_sandbox/self_check.py +124 -16
  394. inspect_ai/util/_subprocess.py +5 -3
  395. inspect_ai/util/_subtask.py +15 -16
  396. {inspect_ai-0.3.62.dist-info → inspect_ai-0.3.64.dist-info}/LICENSE +1 -1
  397. {inspect_ai-0.3.62.dist-info → inspect_ai-0.3.64.dist-info}/METADATA +11 -6
  398. inspect_ai-0.3.64.dist-info/RECORD +625 -0
  399. inspect_ai/_view/www/src/Register.mjs +0 -3
  400. inspect_ai/_view/www/src/Types.mjs +0 -38
  401. inspect_ai/_view/www/src/appearance/Colors.mjs +0 -27
  402. inspect_ai/_view/www/src/appearance/Fonts.mjs +0 -66
  403. inspect_ai/_view/www/src/appearance/Icons.mjs +0 -240
  404. inspect_ai/_view/www/src/components/AnsiDisplay.mjs +0 -184
  405. inspect_ai/_view/www/src/components/AppErrorBoundary.mjs +0 -34
  406. inspect_ai/_view/www/src/components/AsciiCinemaPlayer.mjs +0 -74
  407. inspect_ai/_view/www/src/components/Card.mjs +0 -126
  408. inspect_ai/_view/www/src/components/ChatView.mjs +0 -441
  409. inspect_ai/_view/www/src/components/CopyButton.mjs +0 -48
  410. inspect_ai/_view/www/src/components/Dialog.mjs +0 -61
  411. inspect_ai/_view/www/src/components/DownloadButton.mjs +0 -15
  412. inspect_ai/_view/www/src/components/DownloadPanel.mjs +0 -29
  413. inspect_ai/_view/www/src/components/EmptyPanel.mjs +0 -23
  414. inspect_ai/_view/www/src/components/ErrorPanel.mjs +0 -66
  415. inspect_ai/_view/www/src/components/ExpandablePanel.mjs +0 -136
  416. inspect_ai/_view/www/src/components/FindBand.mjs +0 -157
  417. inspect_ai/_view/www/src/components/HumanBaselineView.mjs +0 -168
  418. inspect_ai/_view/www/src/components/JsonPanel.mjs +0 -61
  419. inspect_ai/_view/www/src/components/LabeledValue.mjs +0 -32
  420. inspect_ai/_view/www/src/components/LargeModal.mjs +0 -190
  421. inspect_ai/_view/www/src/components/LightboxCarousel.mjs +0 -217
  422. inspect_ai/_view/www/src/components/MarkdownDiv.mjs +0 -118
  423. inspect_ai/_view/www/src/components/MessageBand.mjs +0 -48
  424. inspect_ai/_view/www/src/components/MessageContent.mjs +0 -111
  425. inspect_ai/_view/www/src/components/MetaDataGrid.mjs +0 -92
  426. inspect_ai/_view/www/src/components/MetaDataView.mjs +0 -109
  427. inspect_ai/_view/www/src/components/MorePopOver.mjs +0 -50
  428. inspect_ai/_view/www/src/components/NavPills.mjs +0 -63
  429. inspect_ai/_view/www/src/components/ProgressBar.mjs +0 -51
  430. inspect_ai/_view/www/src/components/RenderedContent/ChatMessageRenderer.mjs +0 -54
  431. inspect_ai/_view/www/src/components/RenderedContent/Types.mjs +0 -19
  432. inspect_ai/_view/www/src/components/TabSet.mjs +0 -184
  433. inspect_ai/_view/www/src/components/ToolButton.mjs +0 -16
  434. inspect_ai/_view/www/src/components/Tools.mjs +0 -376
  435. inspect_ai/_view/www/src/components/VirtualList.mjs +0 -280
  436. inspect_ai/_view/www/src/components/ansi-output.js +0 -932
  437. inspect_ai/_view/www/src/json/JsonTab.mjs +0 -48
  438. inspect_ai/_view/www/src/log-reader/Log-Reader.mjs +0 -25
  439. inspect_ai/_view/www/src/log-reader/Native-Log-Reader.mjs +0 -13
  440. inspect_ai/_view/www/src/log-reader/Open-AI-Log-Reader.mjs +0 -263
  441. inspect_ai/_view/www/src/navbar/Navbar.mjs +0 -418
  442. inspect_ai/_view/www/src/navbar/SecondaryBar.mjs +0 -175
  443. inspect_ai/_view/www/src/plan/PlanCard.mjs +0 -418
  444. inspect_ai/_view/www/src/samples/SampleDialog.mjs +0 -123
  445. inspect_ai/_view/www/src/samples/SampleDisplay.mjs +0 -516
  446. inspect_ai/_view/www/src/samples/SampleError.mjs +0 -99
  447. inspect_ai/_view/www/src/samples/SampleList.mjs +0 -427
  448. inspect_ai/_view/www/src/samples/SampleScoreView.mjs +0 -172
  449. inspect_ai/_view/www/src/samples/SampleScores.mjs +0 -34
  450. inspect_ai/_view/www/src/samples/SampleTranscript.mjs +0 -20
  451. inspect_ai/_view/www/src/samples/SamplesDescriptor.mjs +0 -771
  452. inspect_ai/_view/www/src/samples/SamplesTab.mjs +0 -399
  453. inspect_ai/_view/www/src/samples/SamplesTools.mjs +0 -64
  454. inspect_ai/_view/www/src/samples/tools/EpochFilter.mjs +0 -38
  455. inspect_ai/_view/www/src/samples/tools/SampleFilter.mjs +0 -756
  456. inspect_ai/_view/www/src/samples/tools/SelectScorer.mjs +0 -141
  457. inspect_ai/_view/www/src/samples/tools/SortFilter.mjs +0 -151
  458. inspect_ai/_view/www/src/samples/transcript/ApprovalEventView.mjs +0 -71
  459. inspect_ai/_view/www/src/samples/transcript/ErrorEventView.mjs +0 -44
  460. inspect_ai/_view/www/src/samples/transcript/EventPanel.mjs +0 -271
  461. inspect_ai/_view/www/src/samples/transcript/EventRow.mjs +0 -46
  462. inspect_ai/_view/www/src/samples/transcript/EventSection.mjs +0 -33
  463. inspect_ai/_view/www/src/samples/transcript/InfoEventView.mjs +0 -59
  464. inspect_ai/_view/www/src/samples/transcript/InputEventView.mjs +0 -44
  465. inspect_ai/_view/www/src/samples/transcript/LoggerEventView.mjs +0 -32
  466. inspect_ai/_view/www/src/samples/transcript/ModelEventView.mjs +0 -216
  467. inspect_ai/_view/www/src/samples/transcript/SampleInitEventView.mjs +0 -107
  468. inspect_ai/_view/www/src/samples/transcript/SampleLimitEventView.mjs +0 -74
  469. inspect_ai/_view/www/src/samples/transcript/ScoreEventView.mjs +0 -100
  470. inspect_ai/_view/www/src/samples/transcript/StepEventView.mjs +0 -187
  471. inspect_ai/_view/www/src/samples/transcript/SubtaskEventView.mjs +0 -133
  472. inspect_ai/_view/www/src/samples/transcript/ToolEventView.mjs +0 -88
  473. inspect_ai/_view/www/src/samples/transcript/TranscriptView.mjs +0 -459
  474. inspect_ai/_view/www/src/samples/transcript/Types.mjs +0 -44
  475. inspect_ai/_view/www/src/samples/transcript/state/StateDiffView.mjs +0 -53
  476. inspect_ai/_view/www/src/samples/transcript/state/StateEventRenderers.mjs +0 -254
  477. inspect_ai/_view/www/src/samples/transcript/state/StateEventView.mjs +0 -313
  478. inspect_ai/_view/www/src/sidebar/Sidebar.mjs +0 -418
  479. inspect_ai/_view/www/src/usage/ModelTokenTable.mjs +0 -72
  480. inspect_ai/_view/www/src/usage/UsageCard.mjs +0 -159
  481. inspect_ai/_view/www/src/utils/Format.mjs +0 -260
  482. inspect_ai/_view/www/src/utils/Git.mjs +0 -12
  483. inspect_ai/_view/www/src/utils/Html.mjs +0 -21
  484. inspect_ai/_view/www/src/utils/attachments.mjs +0 -31
  485. inspect_ai/_view/www/src/utils/debugging.mjs +0 -23
  486. inspect_ai/_view/www/src/utils/http.mjs +0 -18
  487. inspect_ai/_view/www/src/utils/queue.mjs +0 -67
  488. inspect_ai/_view/www/src/utils/sync.mjs +0 -101
  489. inspect_ai/_view/www/src/workspace/TaskErrorPanel.mjs +0 -17
  490. inspect_ai/_view/www/src/workspace/WorkSpace.mjs +0 -516
  491. inspect_ai/tool/beta/__init__.py +0 -5
  492. inspect_ai-0.3.62.dist-info/RECORD +0 -481
  493. /inspect_ai/{tool/beta/_computer/_resources/tool → _eval}/__init__.py +0 -0
  494. /inspect_ai/{tool/beta/_computer/_resources/tool/requirements.txt → _util/__init__.py} +0 -0
  495. /inspect_ai/_view/www/src/{constants.mjs → constants.ts} +0 -0
  496. /inspect_ai/tool/{beta → _tools}/_computer/__init__.py +0 -0
  497. /inspect_ai/tool/{beta → _tools}/_computer/_computer_split.py +0 -0
  498. /inspect_ai/tool/{beta → _tools}/_computer/_resources/Dockerfile +0 -0
  499. /inspect_ai/tool/{beta → _tools}/_computer/_resources/README.md +0 -0
  500. /inspect_ai/tool/{beta → _tools}/_computer/_resources/entrypoint/entrypoint.sh +0 -0
  501. /inspect_ai/tool/{beta → _tools}/_computer/_resources/entrypoint/novnc_startup.sh +0 -0
  502. /inspect_ai/tool/{beta → _tools}/_computer/_resources/entrypoint/x11vnc_startup.sh +0 -0
  503. /inspect_ai/tool/{beta → _tools}/_computer/_resources/entrypoint/xfce_startup.sh +0 -0
  504. /inspect_ai/tool/{beta → _tools}/_computer/_resources/entrypoint/xvfb_startup.sh +0 -0
  505. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/.config/Code/User/globalStorage/state.vscdb +0 -0
  506. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/.config/Code/User/settings.json +0 -0
  507. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/.config/xfce4/xfconf/xfce-perchannel-xml/xfce4-panel.xml +0 -0
  508. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/.config/xfce4/xfconf/xfce-perchannel-xml/xfce4-screensaver.xml +0 -0
  509. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/Desktop/Firefox Web Browser.desktop +0 -0
  510. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/Desktop/Terminal.desktop +0 -0
  511. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/Desktop/Visual Studio Code.desktop +0 -0
  512. /inspect_ai/tool/{beta → _tools}/_computer/_resources/tool/_logger.py +0 -0
  513. /inspect_ai/tool/{beta → _tools}/_computer/_resources/tool/_run.py +0 -0
  514. /inspect_ai/tool/{beta → _tools}/_computer/_resources/tool/_tool_result.py +0 -0
  515. /inspect_ai/tool/{beta → _tools}/_computer/_resources/tool/computer_tool.py +0 -0
  516. {inspect_ai-0.3.62.dist-info → inspect_ai-0.3.64.dist-info}/WHEEL +0 -0
  517. {inspect_ai-0.3.62.dist-info → inspect_ai-0.3.64.dist-info}/entry_points.txt +0 -0
  518. {inspect_ai-0.3.62.dist-info → inspect_ai-0.3.64.dist-info}/top_level.txt +0 -0
@@ -0,0 +1,331 @@
1
+ import { TabPanel, TabSet } from "../components/TabSet";
2
+ import { MetaDataView } from "../metadata/MetaDataView";
3
+
4
+ import { escapeSelector } from "../utils/html";
5
+ import { isVscode } from "../utils/vscode";
6
+
7
+ import { ApplicationIcons } from "../appearance/icons";
8
+ import { ANSIDisplay } from "../components/AnsiDisplay";
9
+ import { ToolButton } from "../components/ToolButton";
10
+ import { SampleScoreView } from "./scores/SampleScoreView";
11
+
12
+ import clsx from "clsx";
13
+ import { Fragment, MouseEvent, RefObject } from "react";
14
+ import { Card, CardBody, CardHeader } from "../components/Card";
15
+ import { EmptyPanel } from "../components/EmptyPanel";
16
+ import { JSONPanel } from "../components/JsonPanel";
17
+ import {
18
+ kSampleErrorTabId,
19
+ kSampleJsonTabId,
20
+ kSampleMessagesTabId,
21
+ kSampleMetdataTabId,
22
+ kSampleScoringTabId,
23
+ kSampleTranscriptTabId,
24
+ } from "../constants";
25
+ import { EvalSample } from "../types/log";
26
+ import { ModelTokenTable } from "../usage/ModelTokenTable";
27
+ import { printHeadingHtml, printHtml } from "../utils/print";
28
+ import { ChatViewVirtualList } from "./chat/ChatViewVirtualList";
29
+ import { SamplesDescriptor } from "./descriptor/samplesDescriptor";
30
+ import styles from "./SampleDisplay.module.css";
31
+ import { SampleSummaryView } from "./SampleSummaryView";
32
+ import { SampleTranscript } from "./transcript/SampleTranscript";
33
+
34
+ interface SampleDisplayProps {
35
+ id: string;
36
+ sample?: EvalSample;
37
+ sampleDescriptor: SamplesDescriptor;
38
+ selectedTab?: string;
39
+ setSelectedTab: (tab: string) => void;
40
+ scrollRef: RefObject<HTMLDivElement | null>;
41
+ }
42
+
43
+ /**
44
+ * Component to display a sample with relevant context and visibility control.
45
+ */
46
+ export const SampleDisplay: React.FC<SampleDisplayProps> = ({
47
+ id,
48
+ sample,
49
+ sampleDescriptor,
50
+ selectedTab,
51
+ setSelectedTab,
52
+ scrollRef,
53
+ }) => {
54
+ // Tab ids
55
+ const baseId = `sample-dialog`;
56
+
57
+ if (!sample) {
58
+ // Placeholder
59
+ return <EmptyPanel />;
60
+ }
61
+
62
+ // Tab selection
63
+ const onSelectedTab = (e: MouseEvent<HTMLElement>) => {
64
+ const el = e.currentTarget as HTMLElement;
65
+ const id = el.id;
66
+ setSelectedTab(id);
67
+ return false;
68
+ };
69
+
70
+ const scorerNames = Object.keys(sample.scores || {});
71
+ const sampleMetadatas = metadataViewsForSample(`${baseId}-${id}`, sample);
72
+
73
+ const tabsetId = `task-sample-details-tab-${id}`;
74
+ const targetId = `${tabsetId}-content`;
75
+
76
+ const tools = [];
77
+ if (!isVscode()) {
78
+ tools.push(
79
+ <ToolButton
80
+ key="sample-print-tool"
81
+ label="Print"
82
+ icon={ApplicationIcons.copy}
83
+ onClick={() => {
84
+ printSample(id, targetId);
85
+ }}
86
+ />,
87
+ );
88
+ }
89
+
90
+ return (
91
+ <Fragment>
92
+ <SampleSummaryView
93
+ parent_id={id}
94
+ sample={sample}
95
+ sampleDescriptor={sampleDescriptor}
96
+ />
97
+ <TabSet
98
+ id={tabsetId}
99
+ tabControlsClassName={clsx("text-size-base")}
100
+ tabPanelsClassName={clsx(styles.tabPanel)}
101
+ tools={tools}
102
+ >
103
+ {sample.events && sample.events.length > 0 ? (
104
+ <TabPanel
105
+ key={kSampleTranscriptTabId}
106
+ id={kSampleTranscriptTabId}
107
+ className="sample-tab"
108
+ title="Transcript"
109
+ onSelected={onSelectedTab}
110
+ selected={
111
+ selectedTab === kSampleTranscriptTabId ||
112
+ selectedTab === undefined
113
+ }
114
+ scrollable={false}
115
+ >
116
+ <SampleTranscript
117
+ key={`${baseId}-transcript-display-${id}`}
118
+ id={`${baseId}-transcript-display-${id}`}
119
+ evalEvents={sample.events}
120
+ scrollRef={scrollRef}
121
+ />
122
+ </TabPanel>
123
+ ) : null}
124
+ <TabPanel
125
+ key={kSampleMessagesTabId}
126
+ id={kSampleMessagesTabId}
127
+ className={clsx("sample-tab", styles.fullWidth)}
128
+ title="Messages"
129
+ onSelected={onSelectedTab}
130
+ selected={selectedTab === kSampleMessagesTabId}
131
+ scrollable={false}
132
+ >
133
+ <ChatViewVirtualList
134
+ key={`${baseId}-chat-${id}`}
135
+ id={`${baseId}-chat-${id}`}
136
+ messages={sample.messages}
137
+ indented={true}
138
+ scrollRef={scrollRef}
139
+ toolCallStyle="complete"
140
+ />
141
+ </TabPanel>
142
+ {scorerNames.length === 1 ? (
143
+ <TabPanel
144
+ key={kSampleScoringTabId}
145
+ id={kSampleScoringTabId}
146
+ className="sample-tab"
147
+ title="Scoring"
148
+ onSelected={onSelectedTab}
149
+ selected={selectedTab === kSampleScoringTabId}
150
+ >
151
+ <SampleScoreView
152
+ sample={sample}
153
+ sampleDescriptor={sampleDescriptor}
154
+ scorer={scorerNames[0]}
155
+ />
156
+ </TabPanel>
157
+ ) : (
158
+ <>
159
+ {Object.keys(sample.scores || {}).map((scorer) => {
160
+ const tabId = `score-${scorer}`;
161
+ return (
162
+ <TabPanel
163
+ key={tabId}
164
+ id={tabId}
165
+ className="sample-tab"
166
+ title={scorer}
167
+ onSelected={onSelectedTab}
168
+ selected={selectedTab === tabId}
169
+ >
170
+ <SampleScoreView
171
+ sample={sample}
172
+ sampleDescriptor={sampleDescriptor}
173
+ scorer={scorer}
174
+ />
175
+ </TabPanel>
176
+ );
177
+ })}
178
+ </>
179
+ )}
180
+ {sampleMetadatas.length > 0 ? (
181
+ <TabPanel
182
+ id={kSampleMetdataTabId}
183
+ className="sample-tab"
184
+ title="Metadata"
185
+ onSelected={onSelectedTab}
186
+ selected={selectedTab === kSampleMetdataTabId}
187
+ >
188
+ <div className={clsx(styles.metadataPanel)}>{sampleMetadatas}</div>
189
+ </TabPanel>
190
+ ) : null}
191
+ {sample.error ? (
192
+ <TabPanel
193
+ id={kSampleErrorTabId}
194
+ className="sample-tab"
195
+ title="Error"
196
+ onSelected={onSelectedTab}
197
+ selected={selectedTab === kSampleErrorTabId}
198
+ >
199
+ <div className={clsx(styles.padded)}>
200
+ <ANSIDisplay
201
+ output={sample.error.traceback_ansi}
202
+ className={clsx("text-size-small", styles.ansi)}
203
+ />
204
+ </div>
205
+ </TabPanel>
206
+ ) : null}
207
+ {sample.messages.length < 100 ? (
208
+ <TabPanel
209
+ id={kSampleJsonTabId}
210
+ className={"sample-tab"}
211
+ title="JSON"
212
+ onSelected={onSelectedTab}
213
+ selected={selectedTab === kSampleJsonTabId}
214
+ >
215
+ <div className={clsx(styles.padded, styles.fullWidth)}>
216
+ <JSONPanel data={sample} simple={true} />
217
+ </div>
218
+ </TabPanel>
219
+ ) : null}
220
+ </TabSet>
221
+ </Fragment>
222
+ );
223
+ };
224
+
225
+ const metadataViewsForSample = (id: string, sample: EvalSample) => {
226
+ const sampleMetadatas = [];
227
+ if (sample.model_usage && Object.keys(sample.model_usage).length > 0) {
228
+ sampleMetadatas.push(
229
+ <Card key={`sample-usage-${id}`}>
230
+ <CardHeader label="Usage" />
231
+ <CardBody>
232
+ <ModelTokenTable
233
+ model_usage={sample.model_usage}
234
+ className={clsx(styles.noTop)}
235
+ />
236
+ </CardBody>
237
+ </Card>,
238
+ );
239
+ }
240
+
241
+ if (Object.keys(sample?.metadata).length > 0) {
242
+ sampleMetadatas.push(
243
+ <Card key={`sample-metadata-${id}`}>
244
+ <CardHeader label="Metadata" />
245
+ <CardBody>
246
+ <MetaDataView
247
+ id="task-sample-metadata-${id}"
248
+ entries={sample?.metadata as Record<string, unknown>}
249
+ className={clsx("tab-pane", styles.noTop)}
250
+ />
251
+ </CardBody>
252
+ </Card>,
253
+ );
254
+ }
255
+
256
+ if (Object.keys(sample?.store).length > 0) {
257
+ sampleMetadatas.push(
258
+ <Card key={`sample-store-${id}`}>
259
+ <CardHeader label="Store" />
260
+ <CardBody>
261
+ <MetaDataView
262
+ id="task-sample-store-${id}"
263
+ entries={sample?.store as Record<string, unknown>}
264
+ className={clsx("tab-pane", styles.noTop)}
265
+ />
266
+ </CardBody>
267
+ </Card>,
268
+ );
269
+ }
270
+
271
+ return sampleMetadatas;
272
+ };
273
+ const printSample = (id: string, targetId: string) => {
274
+ // The active tab
275
+ const targetTabEl = document.querySelector(
276
+ `#${escapeSelector(targetId)} .sample-tab.tab-pane.show.active`,
277
+ );
278
+ if (targetTabEl) {
279
+ // The target element
280
+ const targetEl = targetTabEl.firstElementChild;
281
+ if (targetEl) {
282
+ // Get the sample heading to include
283
+ const headingId = `sample-heading-${id}`;
284
+ const headingEl = document.getElementById(headingId);
285
+
286
+ // Print the document
287
+ const headingHtml = printHeadingHtml();
288
+ const css = `
289
+ html { font-size: 9pt }
290
+ /* Allow content to break anywhere without any forced page breaks */
291
+ * {
292
+ break-inside: auto; /* Let elements break anywhere */
293
+ page-break-inside: auto; /* Legacy support */
294
+ break-before: auto;
295
+ page-break-before: auto;
296
+ break-after: auto;
297
+ page-break-after: auto;
298
+ }
299
+ /* Specifically disable all page breaks for divs */
300
+ div {
301
+ break-inside: auto;
302
+ page-break-inside: auto;
303
+ }
304
+ body > .transcript-step {
305
+ break-inside: avoid;
306
+ }
307
+ body{
308
+ -webkit-print-color-adjust:exact !important;
309
+ print-color-adjust:exact !important;
310
+ }
311
+ /* Allow preformatted text and code blocks to break across pages */
312
+ pre, code {
313
+ white-space: pre-wrap; /* Wrap long lines instead of keeping them on one line */
314
+ overflow-wrap: break-word; /* Ensure long words are broken to fit within the page */
315
+ break-inside: auto; /* Allow page breaks inside the element */
316
+ page-break-inside: auto; /* Older equivalent */
317
+ }
318
+
319
+ /* Additional control for long lines within code/preformatted blocks */
320
+ pre {
321
+ word-wrap: break-word; /* Break long words if needed */
322
+ }
323
+
324
+ `;
325
+ printHtml(
326
+ [headingHtml, headingEl?.outerHTML, targetEl.innerHTML].join("\n"),
327
+ css,
328
+ );
329
+ }
330
+ }
331
+ };
@@ -0,0 +1,24 @@
1
+ .target {
2
+ padding-left: 0;
3
+ }
4
+
5
+ .answer {
6
+ padding-left: 0;
7
+ }
8
+
9
+ .grid {
10
+ display: grid;
11
+ grid-column-gap: 0.5em;
12
+ border-bottom: solid var(--bs-border-color) 1px;
13
+ margin-bottom: 1em;
14
+ padding: 0em 1em 1em 1em;
15
+ }
16
+
17
+ .centerLabel {
18
+ display: flex;
19
+ justify-content: center;
20
+ }
21
+
22
+ .wrap {
23
+ word-wrap: anywhere;
24
+ }
@@ -0,0 +1,177 @@
1
+ import clsx from "clsx";
2
+ import { MarkdownDiv } from "../components/MarkdownDiv";
3
+ import { EvalSample } from "../types/log";
4
+ import { arrayToString, inputString } from "../utils/format";
5
+ import { SamplesDescriptor } from "./descriptor/samplesDescriptor";
6
+ import { FlatSampleError } from "./error/FlatSampleErrorView";
7
+
8
+ import { ReactNode } from "react";
9
+ import styles from "./SampleSummaryView.module.css";
10
+
11
+ interface SampleSummaryViewProps {
12
+ parent_id: string;
13
+ sample: EvalSample;
14
+ sampleDescriptor: SamplesDescriptor;
15
+ }
16
+
17
+ interface SummaryColumn {
18
+ label: string;
19
+ value: string | ReactNode;
20
+ size: string;
21
+ center?: boolean;
22
+ clamp?: boolean;
23
+ }
24
+
25
+ /**
26
+ * Component to display a sample with relevant context and visibility control.
27
+ */
28
+ export const SampleSummaryView: React.FC<SampleSummaryViewProps> = ({
29
+ parent_id,
30
+ sample,
31
+ sampleDescriptor,
32
+ }) => {
33
+ const input =
34
+ sampleDescriptor?.messageShape.normalized.input > 0
35
+ ? Math.max(0.15, sampleDescriptor.messageShape.normalized.input)
36
+ : 0;
37
+ const target =
38
+ sampleDescriptor?.messageShape.normalized.target > 0
39
+ ? Math.max(0.15, sampleDescriptor.messageShape.normalized.target)
40
+ : 0;
41
+ const answer =
42
+ sampleDescriptor?.messageShape.normalized.answer > 0
43
+ ? Math.max(0.15, sampleDescriptor.messageShape.normalized.answer)
44
+ : 0;
45
+ const limitSize =
46
+ sampleDescriptor?.messageShape.normalized.limit > 0
47
+ ? Math.max(0.15, sampleDescriptor.messageShape.normalized.limit)
48
+ : 0;
49
+ const idSize = Math.max(
50
+ 2,
51
+ Math.min(10, sampleDescriptor?.messageShape.raw.id),
52
+ );
53
+
54
+ const scoreInput = inputString(sample.input);
55
+ if (sample.choices && sample.choices.length > 0) {
56
+ scoreInput.push("");
57
+ scoreInput.push(
58
+ ...sample.choices.map((choice, index) => {
59
+ return `${String.fromCharCode(65 + index)}) ${choice}`;
60
+ }),
61
+ );
62
+ }
63
+
64
+ // The columns for the sample
65
+ const columns: SummaryColumn[] = [];
66
+ columns.push({
67
+ label: "Id",
68
+ value: sample.id,
69
+ size: `${idSize}em`,
70
+ });
71
+
72
+ columns.push({
73
+ label: "Input",
74
+ value: scoreInput,
75
+ size: `${input}fr`,
76
+ clamp: true,
77
+ });
78
+
79
+ if (sample.target) {
80
+ columns.push({
81
+ label: "Target",
82
+ value: (
83
+ <MarkdownDiv
84
+ markdown={arrayToString(arrayToString(sample?.target || "none"))}
85
+ className={clsx("no-last-para-padding", styles.target)}
86
+ />
87
+ ),
88
+ size: `${target}fr`,
89
+ clamp: true,
90
+ });
91
+ }
92
+
93
+ const fullAnswer =
94
+ sample && sampleDescriptor
95
+ ? sampleDescriptor.selectedScorerDescriptor(sample).answer()
96
+ : undefined;
97
+ if (fullAnswer) {
98
+ columns.push({
99
+ label: "Answer",
100
+ value: sample ? (
101
+ <MarkdownDiv
102
+ markdown={fullAnswer}
103
+ className={clsx("no-last-para-padding", styles.answer)}
104
+ />
105
+ ) : (
106
+ ""
107
+ ),
108
+ size: `${answer}fr`,
109
+ clamp: true,
110
+ });
111
+ }
112
+
113
+ if (sample?.limit && limitSize > 0) {
114
+ columns.push({
115
+ label: "Limit",
116
+ value: sample.limit.type,
117
+ size: `${limitSize}fr`,
118
+ center: true,
119
+ });
120
+ }
121
+
122
+ columns.push({
123
+ label: "Score",
124
+ value: sample.error ? (
125
+ <FlatSampleError message={sample.error.message} />
126
+ ) : (
127
+ // TODO: Cleanup once the PR lands which makes sample / sample summary share common interface
128
+ sampleDescriptor?.selectedScore(sample)?.render() || ""
129
+ ),
130
+ size: "minmax(2em, auto)",
131
+ center: true,
132
+ });
133
+
134
+ return (
135
+ <div
136
+ id={`sample-heading-${parent_id}`}
137
+ className={clsx(styles.grid, "text-size-base")}
138
+ style={{
139
+ gridTemplateColumns: `${columns
140
+ .map((col) => {
141
+ return col.size;
142
+ })
143
+ .join(" ")}`,
144
+ }}
145
+ >
146
+ {columns.map((col, idx) => {
147
+ return (
148
+ <div
149
+ key={`sample-summ-lbl-${idx}`}
150
+ className={clsx(
151
+ "text-style-label",
152
+ "text-style-secondary",
153
+ "text-size-base",
154
+ col.center ? styles.centerLabel : undefined,
155
+ )}
156
+ >
157
+ {col.label}
158
+ </div>
159
+ );
160
+ })}
161
+ {columns.map((col, idx) => {
162
+ return (
163
+ <div
164
+ key={`sample-summ-val-${idx}`}
165
+ className={clsx(
166
+ styles.wrap,
167
+ col.clamp ? "three-line-clamp" : undefined,
168
+ col.center ? styles.centerLabel : undefined,
169
+ )}
170
+ >
171
+ {col.value}
172
+ </div>
173
+ );
174
+ })}
175
+ </div>
176
+ );
177
+ };
@@ -0,0 +1,52 @@
1
+ import { Fragment } from "react/jsx-runtime";
2
+ import { ScoreFilter, ScoreLabel } from "../types";
3
+ import { SamplesDescriptor } from "./descriptor/samplesDescriptor";
4
+ import { EpochFilter } from "./sample-tools/EpochFilter";
5
+ import { SampleFilter } from "./sample-tools/sample-filter/SampleFilter";
6
+ import { SelectScorer } from "./sample-tools/SelectScorer";
7
+ import { SortFilter } from "./sample-tools/SortFilter";
8
+
9
+ interface SampleToolsProps {
10
+ epoch: string;
11
+ setEpoch: (epoch: string) => void;
12
+ epochs: number;
13
+ scoreFilter: ScoreFilter;
14
+ setScoreFilter: (filter: ScoreFilter) => void;
15
+ sort: string;
16
+ setSort: (sort: string) => void;
17
+ score?: ScoreLabel;
18
+ setScore: (score: ScoreLabel) => void;
19
+ scores: ScoreLabel[];
20
+ sampleDescriptor: SamplesDescriptor;
21
+ }
22
+
23
+ export const SampleTools: React.FC<SampleToolsProps> = ({
24
+ epoch,
25
+ setEpoch,
26
+ epochs,
27
+ scoreFilter,
28
+ setScoreFilter,
29
+ sort,
30
+ setSort,
31
+ score,
32
+ setScore,
33
+ scores,
34
+ sampleDescriptor,
35
+ }) => {
36
+ return (
37
+ <Fragment>
38
+ <SampleFilter
39
+ evalDescriptor={sampleDescriptor.evalDescriptor}
40
+ scoreFilter={scoreFilter}
41
+ setScoreFilter={setScoreFilter}
42
+ />
43
+ {scores.length > 1 ? (
44
+ <SelectScorer scores={scores} score={score} setScore={setScore} />
45
+ ) : undefined}
46
+ {epochs > 1 ? (
47
+ <EpochFilter epoch={epoch} setEpoch={setEpoch} epochs={epochs} />
48
+ ) : undefined}
49
+ <SortFilter sort={sort} setSort={setSort} epochs={epochs} />
50
+ </Fragment>
51
+ );
52
+ };
@@ -0,0 +1,29 @@
1
+ .message {
2
+ font-weight: 300;
3
+ padding-bottom: 0.5em;
4
+ margin-left: 0;
5
+ margin-right: 0;
6
+ white-space: normal;
7
+ }
8
+
9
+ .systemRole {
10
+ opacity: 0.7;
11
+ }
12
+
13
+ .messageGrid {
14
+ display: grid;
15
+ grid-template-columns: max-content auto;
16
+ column-gap: 0.3em;
17
+ font-weight: 500;
18
+ margin-bottom: 0.5em;
19
+ }
20
+
21
+ .messageContents {
22
+ margin-left: 0;
23
+ padding-bottom: 0;
24
+ }
25
+
26
+ .messageContents.indented {
27
+ margin-left: 1.1rem;
28
+ padding-bottom: 0.8rem;
29
+ }
@@ -0,0 +1,76 @@
1
+ import clsx from "clsx";
2
+ import { Fragment } from "react";
3
+ import ExpandablePanel from "../../components/ExpandablePanel";
4
+ import { MarkdownDiv } from "../../components/MarkdownDiv";
5
+ import {
6
+ ChatMessageAssistant,
7
+ ChatMessageSystem,
8
+ ChatMessageTool,
9
+ ChatMessageUser,
10
+ } from "../../types/log";
11
+ import styles from "./ChatMessage.module.css";
12
+ import { MessageContents } from "./MessageContents";
13
+ import { iconForMsg } from "./messages";
14
+
15
+ interface ChatMessageProps {
16
+ id: string;
17
+ message: ChatMessageAssistant | ChatMessageSystem | ChatMessageUser;
18
+ toolMessages: ChatMessageTool[];
19
+ indented?: boolean;
20
+ toolCallStyle: "compact" | "complete";
21
+ }
22
+
23
+ export const ChatMessage: React.FC<ChatMessageProps> = ({
24
+ id,
25
+ message,
26
+ toolMessages,
27
+ indented,
28
+ toolCallStyle,
29
+ }) => {
30
+ const collapse = message.role === "system";
31
+ return (
32
+ <div
33
+ className={clsx(
34
+ message.role,
35
+ "text-size-base",
36
+ styles.message,
37
+ message.role === "system" ? styles.systemRole : undefined,
38
+ )}
39
+ >
40
+ <div className={clsx(styles.messageGrid, "text-style-label")}>
41
+ <i className={iconForMsg(message)} />
42
+ {message.role}
43
+ </div>
44
+ {message.role === "assistant" && message.reasoning ? (
45
+ <Fragment key={`${id}-response-label`}>
46
+ <div className={clsx("text-style-label", "text-style-secondary")}>
47
+ Reasoning
48
+ </div>
49
+ <ExpandablePanel collapse={true}>
50
+ <MarkdownDiv markdown={message.reasoning} />
51
+ </ExpandablePanel>
52
+ </Fragment>
53
+ ) : undefined}
54
+ <div
55
+ className={clsx(
56
+ styles.messageContents,
57
+ indented ? styles.indented : undefined,
58
+ )}
59
+ >
60
+ {message.role === "assistant" && message.reasoning ? (
61
+ <div className={clsx("text-style-label", "text-style-secondary")}>
62
+ Response
63
+ </div>
64
+ ) : undefined}
65
+ <ExpandablePanel collapse={collapse}>
66
+ <MessageContents
67
+ key={`${id}-contents`}
68
+ message={message}
69
+ toolMessages={toolMessages}
70
+ toolCallStyle={toolCallStyle}
71
+ />
72
+ </ExpandablePanel>
73
+ </div>
74
+ </div>
75
+ );
76
+ };