inspect-ai 0.3.62__py3-none-any.whl → 0.3.63__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 (415) hide show
  1. inspect_ai/_cli/main.py +1 -1
  2. inspect_ai/_cli/trace.py +8 -0
  3. inspect_ai/_display/core/active.py +2 -3
  4. inspect_ai/_eval/eval.py +4 -4
  5. inspect_ai/_eval/evalset.py +6 -6
  6. inspect_ai/_eval/task/run.py +3 -0
  7. inspect_ai/_util/hash.py +1 -1
  8. inspect_ai/_view/www/.vscode/extensions.json +3 -0
  9. inspect_ai/_view/www/.vscode/settings.json +8 -0
  10. inspect_ai/_view/www/App.css +92 -29
  11. inspect_ai/_view/www/dist/assets/index.css +16637 -14676
  12. inspect_ai/_view/www/dist/assets/index.js +58897 -51440
  13. inspect_ai/_view/www/dist/index.html +1 -1
  14. inspect_ai/_view/www/index.html +2 -2
  15. inspect_ai/_view/www/log-schema.json +1 -0
  16. inspect_ai/_view/www/package.json +22 -4
  17. inspect_ai/_view/www/postcss.config.cjs +8 -9
  18. inspect_ai/_view/www/src/{App.mjs → App.tsx} +355 -365
  19. inspect_ai/_view/www/src/AppErrorBoundary.tsx +47 -0
  20. inspect_ai/_view/www/src/api/api-browser.ts +2 -2
  21. inspect_ai/_view/www/src/api/api-http.ts +3 -5
  22. inspect_ai/_view/www/src/api/api-vscode.ts +6 -6
  23. inspect_ai/_view/www/src/api/client-api.ts +4 -4
  24. inspect_ai/_view/www/src/api/index.ts +4 -4
  25. inspect_ai/_view/www/src/api/{Types.ts → types.ts} +25 -9
  26. inspect_ai/_view/www/src/appearance/colors.ts +9 -0
  27. inspect_ai/_view/www/src/appearance/fonts.ts +39 -0
  28. inspect_ai/_view/www/src/appearance/icons.ts +100 -0
  29. inspect_ai/_view/www/src/appearance/{Styles.mjs → styles.ts} +2 -32
  30. inspect_ai/_view/www/src/components/AnsiDisplay.tsx +198 -0
  31. inspect_ai/_view/www/src/components/AsciinemaPlayer.tsx +86 -0
  32. inspect_ai/_view/www/src/components/Card.css +60 -0
  33. inspect_ai/_view/www/src/components/Card.tsx +109 -0
  34. inspect_ai/_view/www/src/components/CopyButton.module.css +11 -0
  35. inspect_ai/_view/www/src/components/CopyButton.tsx +58 -0
  36. inspect_ai/_view/www/src/components/DownloadButton.css +4 -0
  37. inspect_ai/_view/www/src/components/DownloadButton.tsx +25 -0
  38. inspect_ai/_view/www/src/components/DownloadPanel.css +10 -0
  39. inspect_ai/_view/www/src/components/DownloadPanel.tsx +30 -0
  40. inspect_ai/_view/www/src/components/EmptyPanel.css +12 -0
  41. inspect_ai/_view/www/src/components/EmptyPanel.tsx +15 -0
  42. inspect_ai/_view/www/src/components/ErrorPanel.css +37 -0
  43. inspect_ai/_view/www/src/components/ErrorPanel.tsx +39 -0
  44. inspect_ai/_view/www/src/components/ExpandablePanel.css +40 -0
  45. inspect_ai/_view/www/src/components/ExpandablePanel.tsx +115 -0
  46. inspect_ai/_view/www/src/components/FindBand.css +49 -0
  47. inspect_ai/_view/www/src/components/FindBand.tsx +130 -0
  48. inspect_ai/_view/www/src/components/HumanBaselineView.css +41 -0
  49. inspect_ai/_view/www/src/components/HumanBaselineView.tsx +162 -0
  50. inspect_ai/_view/www/src/components/JsonPanel.css +20 -0
  51. inspect_ai/_view/www/src/components/JsonPanel.tsx +82 -0
  52. inspect_ai/_view/www/src/components/LabeledValue.css +20 -0
  53. inspect_ai/_view/www/src/components/LabeledValue.tsx +41 -0
  54. inspect_ai/_view/www/src/components/LargeModal.module.css +54 -0
  55. inspect_ai/_view/www/src/components/LargeModal.tsx +199 -0
  56. inspect_ai/_view/www/src/components/LightboxCarousel.css +95 -0
  57. inspect_ai/_view/www/src/components/LightboxCarousel.tsx +132 -0
  58. inspect_ai/_view/www/src/components/MarkdownDiv.css +3 -0
  59. inspect_ai/_view/www/src/components/MarkdownDiv.tsx +133 -0
  60. inspect_ai/_view/www/src/components/MessageBand.css +43 -0
  61. inspect_ai/_view/www/src/components/MessageBand.tsx +39 -0
  62. inspect_ai/_view/www/src/components/MorePopOver.tsx +67 -0
  63. inspect_ai/_view/www/src/components/NavPills.module.css +18 -0
  64. inspect_ai/_view/www/src/components/NavPills.tsx +99 -0
  65. inspect_ai/_view/www/src/components/ProgressBar.module.css +37 -0
  66. inspect_ai/_view/www/src/components/ProgressBar.tsx +22 -0
  67. inspect_ai/_view/www/src/components/TabSet.module.css +40 -0
  68. inspect_ai/_view/www/src/components/TabSet.tsx +200 -0
  69. inspect_ai/_view/www/src/components/ToolButton.css +3 -0
  70. inspect_ai/_view/www/src/components/ToolButton.tsx +27 -0
  71. inspect_ai/_view/www/src/components/VirtualList.module.css +19 -0
  72. inspect_ai/_view/www/src/components/VirtualList.tsx +292 -0
  73. inspect_ai/_view/www/src/{index.js → index.tsx} +45 -19
  74. inspect_ai/_view/www/src/{log → logfile}/remoteLogFile.ts +3 -7
  75. inspect_ai/_view/www/src/{utils/remoteZipFile.mjs → logfile/remoteZipFile.ts} +86 -80
  76. inspect_ai/_view/www/src/metadata/MetaDataGrid.tsx +83 -0
  77. inspect_ai/_view/www/src/metadata/MetaDataView.module.css +35 -0
  78. inspect_ai/_view/www/src/metadata/MetaDataView.tsx +95 -0
  79. inspect_ai/_view/www/src/metadata/MetadataGrid.module.css +15 -0
  80. inspect_ai/_view/www/src/metadata/RenderedContent.module.css +12 -0
  81. inspect_ai/_view/www/src/{components/RenderedContent/RenderedContent.mjs → metadata/RenderedContent.tsx} +92 -73
  82. inspect_ai/_view/www/src/metadata/types.ts +18 -0
  83. inspect_ai/_view/www/src/plan/DatasetDetailView.module.css +3 -0
  84. inspect_ai/_view/www/src/plan/DatasetDetailView.tsx +37 -0
  85. inspect_ai/_view/www/src/plan/DetailStep.module.css +9 -0
  86. inspect_ai/_view/www/src/plan/DetailStep.tsx +31 -0
  87. inspect_ai/_view/www/src/plan/PlanCard.tsx +28 -0
  88. inspect_ai/_view/www/src/plan/PlanDetailView.module.css +48 -0
  89. inspect_ai/_view/www/src/plan/PlanDetailView.tsx +309 -0
  90. inspect_ai/_view/www/src/plan/ScorerDetailView.module.css +3 -0
  91. inspect_ai/_view/www/src/plan/ScorerDetailView.tsx +30 -0
  92. inspect_ai/_view/www/src/plan/SolverDetailView.module.css +15 -0
  93. inspect_ai/_view/www/src/plan/SolverDetailView.tsx +32 -0
  94. inspect_ai/_view/www/src/samples/InlineSampleDisplay.module.css +8 -0
  95. inspect_ai/_view/www/src/samples/InlineSampleDisplay.tsx +53 -0
  96. inspect_ai/_view/www/src/samples/SampleDialog.tsx +122 -0
  97. inspect_ai/_view/www/src/samples/SampleDisplay.module.css +29 -0
  98. inspect_ai/_view/www/src/samples/SampleDisplay.tsx +326 -0
  99. inspect_ai/_view/www/src/samples/SampleSummaryView.module.css +24 -0
  100. inspect_ai/_view/www/src/samples/SampleSummaryView.tsx +175 -0
  101. inspect_ai/_view/www/src/samples/SamplesTools.tsx +60 -0
  102. inspect_ai/_view/www/src/samples/chat/ChatMessage.module.css +29 -0
  103. inspect_ai/_view/www/src/samples/chat/ChatMessage.tsx +76 -0
  104. inspect_ai/_view/www/src/samples/chat/ChatMessageRenderer.tsx +60 -0
  105. inspect_ai/_view/www/src/samples/chat/ChatMessageRow.module.css +9 -0
  106. inspect_ai/_view/www/src/samples/chat/ChatMessageRow.tsx +57 -0
  107. inspect_ai/_view/www/src/samples/chat/ChatView.tsx +46 -0
  108. inspect_ai/_view/www/src/samples/chat/ChatViewVirtualList.module.css +4 -0
  109. inspect_ai/_view/www/src/samples/chat/ChatViewVirtualList.tsx +58 -0
  110. inspect_ai/_view/www/src/samples/chat/MessageContent.module.css +4 -0
  111. inspect_ai/_view/www/src/samples/chat/MessageContent.tsx +143 -0
  112. inspect_ai/_view/www/src/samples/chat/MessageContents.module.css +3 -0
  113. inspect_ai/_view/www/src/samples/chat/MessageContents.tsx +131 -0
  114. inspect_ai/_view/www/src/samples/chat/messages.ts +112 -0
  115. inspect_ai/_view/www/src/samples/chat/tools/ToolCallView.tsx +145 -0
  116. inspect_ai/_view/www/src/samples/chat/tools/ToolInput.module.css +14 -0
  117. inspect_ai/_view/www/src/samples/chat/tools/ToolInput.tsx +86 -0
  118. inspect_ai/_view/www/src/samples/chat/tools/ToolOutput.module.css +19 -0
  119. inspect_ai/_view/www/src/samples/chat/tools/ToolOutput.tsx +53 -0
  120. inspect_ai/_view/www/src/samples/chat/tools/ToolTitle.module.css +4 -0
  121. inspect_ai/_view/www/src/samples/chat/tools/ToolTitle.tsx +18 -0
  122. inspect_ai/_view/www/src/samples/chat/tools/tool.ts +107 -0
  123. inspect_ai/_view/www/src/samples/descriptor/samplesDescriptor.tsx +363 -0
  124. inspect_ai/_view/www/src/samples/descriptor/score/BooleanScoreDescriptor.module.css +22 -0
  125. inspect_ai/_view/www/src/samples/descriptor/score/BooleanScoreDescriptor.tsx +26 -0
  126. inspect_ai/_view/www/src/samples/descriptor/score/CategoricalScoreDescriptor.tsx +18 -0
  127. inspect_ai/_view/www/src/samples/descriptor/score/NumericScoreDescriptor.tsx +27 -0
  128. inspect_ai/_view/www/src/samples/descriptor/score/ObjectScoreDescriptor.module.css +18 -0
  129. inspect_ai/_view/www/src/samples/descriptor/score/ObjectScoreDescriptor.tsx +71 -0
  130. inspect_ai/_view/www/src/samples/descriptor/score/OtherScoreDescriptor.tsx +20 -0
  131. inspect_ai/_view/www/src/samples/descriptor/score/PassFailScoreDescriptor.module.css +28 -0
  132. inspect_ai/_view/www/src/samples/descriptor/score/PassFailScoreDescriptor.tsx +81 -0
  133. inspect_ai/_view/www/src/samples/descriptor/score/ScoreDescriptor.tsx +99 -0
  134. inspect_ai/_view/www/src/samples/descriptor/types.ts +55 -0
  135. inspect_ai/_view/www/src/samples/error/FlatSampleErrorView.module.css +19 -0
  136. inspect_ai/_view/www/src/samples/error/FlatSampleErrorView.tsx +22 -0
  137. inspect_ai/_view/www/src/samples/error/SampleErrorView.module.css +17 -0
  138. inspect_ai/_view/www/src/samples/error/SampleErrorView.tsx +31 -0
  139. inspect_ai/_view/www/src/samples/error/error.ts +15 -0
  140. inspect_ai/_view/www/src/samples/list/SampleFooter.module.css +9 -0
  141. inspect_ai/_view/www/src/samples/list/SampleFooter.tsx +14 -0
  142. inspect_ai/_view/www/src/samples/list/SampleHeader.module.css +13 -0
  143. inspect_ai/_view/www/src/samples/list/SampleHeader.tsx +36 -0
  144. inspect_ai/_view/www/src/samples/list/SampleList.module.css +11 -0
  145. inspect_ai/_view/www/src/samples/list/SampleList.tsx +247 -0
  146. inspect_ai/_view/www/src/samples/list/SampleRow.module.css +33 -0
  147. inspect_ai/_view/www/src/samples/list/SampleRow.tsx +98 -0
  148. inspect_ai/_view/www/src/samples/list/SampleSeparator.module.css +6 -0
  149. inspect_ai/_view/www/src/samples/list/SampleSeparator.tsx +24 -0
  150. inspect_ai/_view/www/src/samples/sample-tools/EpochFilter.module.css +9 -0
  151. inspect_ai/_view/www/src/samples/sample-tools/EpochFilter.tsx +51 -0
  152. inspect_ai/_view/www/src/samples/sample-tools/SelectScorer.module.css +16 -0
  153. inspect_ai/_view/www/src/samples/sample-tools/SelectScorer.tsx +173 -0
  154. inspect_ai/_view/www/src/samples/sample-tools/SortFilter.module.css +9 -0
  155. inspect_ai/_view/www/src/samples/sample-tools/SortFilter.tsx +182 -0
  156. inspect_ai/_view/www/src/samples/{tools/filters.mjs → sample-tools/filters.ts} +86 -81
  157. inspect_ai/_view/www/src/samples/sample-tools/sample-filter/SampleFilter.module.css +16 -0
  158. inspect_ai/_view/www/src/samples/sample-tools/sample-filter/SampleFilter.tsx +288 -0
  159. inspect_ai/_view/www/src/samples/sample-tools/sample-filter/completions.ts +346 -0
  160. inspect_ai/_view/www/src/samples/sample-tools/sample-filter/language.ts +19 -0
  161. inspect_ai/_view/www/src/samples/sample-tools/sample-filter/tokenize.ts +97 -0
  162. inspect_ai/_view/www/src/samples/{SampleLimit.mjs → sampleLimit.ts} +3 -6
  163. inspect_ai/_view/www/src/samples/scores/SampleScoreView.module.css +53 -0
  164. inspect_ai/_view/www/src/samples/scores/SampleScoreView.tsx +168 -0
  165. inspect_ai/_view/www/src/samples/scores/SampleScores.module.css +5 -0
  166. inspect_ai/_view/www/src/samples/scores/SampleScores.tsx +37 -0
  167. inspect_ai/_view/www/src/samples/transcript/ApprovalEventView.tsx +66 -0
  168. inspect_ai/_view/www/src/samples/transcript/ErrorEventView.tsx +51 -0
  169. inspect_ai/_view/www/src/samples/transcript/InfoEventView.module.css +3 -0
  170. inspect_ai/_view/www/src/samples/transcript/InfoEventView.tsx +54 -0
  171. inspect_ai/_view/www/src/samples/transcript/InputEventView.tsx +48 -0
  172. inspect_ai/_view/www/src/samples/transcript/LoggerEventView.module.css +6 -0
  173. inspect_ai/_view/www/src/samples/transcript/LoggerEventView.tsx +36 -0
  174. inspect_ai/_view/www/src/samples/transcript/ModelEventView.module.css +43 -0
  175. inspect_ai/_view/www/src/samples/transcript/ModelEventView.tsx +223 -0
  176. inspect_ai/_view/www/src/samples/transcript/SampleInitEventView.module.css +23 -0
  177. inspect_ai/_view/www/src/samples/transcript/SampleInitEventView.tsx +108 -0
  178. inspect_ai/_view/www/src/samples/transcript/SampleLimitEventView.tsx +75 -0
  179. inspect_ai/_view/www/src/samples/transcript/SampleTranscript.tsx +22 -0
  180. inspect_ai/_view/www/src/samples/transcript/ScoreEventView.module.css +15 -0
  181. inspect_ai/_view/www/src/samples/transcript/ScoreEventView.tsx +100 -0
  182. inspect_ai/_view/www/src/samples/transcript/StepEventView.tsx +171 -0
  183. inspect_ai/_view/www/src/samples/transcript/SubtaskEventView.module.css +19 -0
  184. inspect_ai/_view/www/src/samples/transcript/SubtaskEventView.tsx +133 -0
  185. inspect_ai/_view/www/src/samples/transcript/ToolEventView.module.css +10 -0
  186. inspect_ai/_view/www/src/samples/transcript/ToolEventView.tsx +91 -0
  187. inspect_ai/_view/www/src/samples/transcript/TranscriptView.module.css +49 -0
  188. inspect_ai/_view/www/src/samples/transcript/TranscriptView.tsx +449 -0
  189. inspect_ai/_view/www/src/samples/transcript/event/EventNav.module.css +5 -0
  190. inspect_ai/_view/www/src/samples/transcript/event/EventNav.tsx +43 -0
  191. inspect_ai/_view/www/src/samples/transcript/event/EventNavs.module.css +3 -0
  192. inspect_ai/_view/www/src/samples/transcript/event/EventNavs.tsx +38 -0
  193. inspect_ai/_view/www/src/samples/transcript/event/EventPanel.module.css +25 -0
  194. inspect_ai/_view/www/src/samples/transcript/event/EventPanel.tsx +190 -0
  195. inspect_ai/_view/www/src/samples/transcript/event/EventRow.module.css +13 -0
  196. inspect_ai/_view/www/src/samples/transcript/event/EventRow.tsx +32 -0
  197. inspect_ai/_view/www/src/samples/transcript/event/EventSection.module.css +8 -0
  198. inspect_ai/_view/www/src/samples/transcript/event/EventSection.tsx +29 -0
  199. inspect_ai/_view/www/src/samples/transcript/state/StateDiffView.tsx +67 -0
  200. inspect_ai/_view/www/src/samples/transcript/state/StateEventRenderers.tsx +274 -0
  201. inspect_ai/_view/www/src/samples/transcript/state/StateEventRenders.module.css +10 -0
  202. inspect_ai/_view/www/src/samples/transcript/state/StateEventView.module.css +9 -0
  203. inspect_ai/_view/www/src/samples/transcript/state/{StateEventView.mjs → StateEventView.tsx} +148 -110
  204. inspect_ai/_view/www/src/samples/transcript/types.ts +58 -0
  205. inspect_ai/_view/www/src/types/log.d.ts +1 -0
  206. inspect_ai/_view/www/src/types/prism.d.ts +11 -0
  207. inspect_ai/_view/www/src/types.ts +71 -0
  208. inspect_ai/_view/www/src/usage/ModelTokenTable.tsx +22 -0
  209. inspect_ai/_view/www/src/usage/ModelUsagePanel.module.css +24 -0
  210. inspect_ai/_view/www/src/usage/ModelUsagePanel.tsx +95 -0
  211. inspect_ai/_view/www/src/usage/TokenTable.module.css +17 -0
  212. inspect_ai/_view/www/src/usage/TokenTable.tsx +91 -0
  213. inspect_ai/_view/www/src/usage/UsageCard.module.css +15 -0
  214. inspect_ai/_view/www/src/usage/UsageCard.tsx +67 -0
  215. inspect_ai/_view/www/src/utils/attachments.ts +42 -0
  216. inspect_ai/_view/www/src/utils/{Base64.mjs → base64.ts} +1 -6
  217. inspect_ai/_view/www/src/{components/Browser.mjs → utils/browser.ts} +0 -1
  218. inspect_ai/_view/www/src/utils/debugging.ts +28 -0
  219. inspect_ai/_view/www/src/utils/dom.ts +30 -0
  220. inspect_ai/_view/www/src/utils/format.ts +194 -0
  221. inspect_ai/_view/www/src/utils/git.ts +7 -0
  222. inspect_ai/_view/www/src/utils/html.ts +6 -0
  223. inspect_ai/_view/www/src/utils/http.ts +14 -0
  224. inspect_ai/_view/www/src/utils/{Path.mjs → path.ts} +2 -9
  225. inspect_ai/_view/www/src/utils/{Print.mjs → print.ts} +34 -26
  226. inspect_ai/_view/www/src/utils/queue.ts +51 -0
  227. inspect_ai/_view/www/src/utils/sync.ts +114 -0
  228. inspect_ai/_view/www/src/utils/{Type.mjs → type.ts} +3 -6
  229. inspect_ai/_view/www/src/utils/vscode.ts +13 -0
  230. inspect_ai/_view/www/src/workspace/WorkSpace.tsx +324 -0
  231. inspect_ai/_view/www/src/workspace/WorkSpaceView.module.css +33 -0
  232. inspect_ai/_view/www/src/workspace/WorkSpaceView.tsx +160 -0
  233. inspect_ai/_view/www/src/workspace/error/TaskErrorPanel.module.css +3 -0
  234. inspect_ai/_view/www/src/workspace/error/TaskErrorPanel.tsx +28 -0
  235. inspect_ai/_view/www/src/workspace/navbar/Navbar.module.css +54 -0
  236. inspect_ai/_view/www/src/workspace/navbar/Navbar.tsx +68 -0
  237. inspect_ai/_view/www/src/workspace/navbar/PrimaryBar.module.css +52 -0
  238. inspect_ai/_view/www/src/workspace/navbar/PrimaryBar.tsx +113 -0
  239. inspect_ai/_view/www/src/workspace/navbar/ResultsPanel.module.css +67 -0
  240. inspect_ai/_view/www/src/workspace/navbar/ResultsPanel.tsx +156 -0
  241. inspect_ai/_view/www/src/workspace/navbar/SecondaryBar.module.css +28 -0
  242. inspect_ai/_view/www/src/workspace/navbar/SecondaryBar.tsx +222 -0
  243. inspect_ai/_view/www/src/workspace/navbar/StatusPanel.module.css +14 -0
  244. inspect_ai/_view/www/src/workspace/navbar/StatusPanel.tsx +61 -0
  245. inspect_ai/_view/www/src/workspace/sidebar/EvalStatus.module.css +15 -0
  246. inspect_ai/_view/www/src/workspace/sidebar/EvalStatus.tsx +71 -0
  247. inspect_ai/_view/www/src/workspace/sidebar/LogDirectoryTitleView.module.css +5 -0
  248. inspect_ai/_view/www/src/workspace/sidebar/LogDirectoryTitleView.tsx +56 -0
  249. inspect_ai/_view/www/src/workspace/sidebar/Sidebar.module.css +68 -0
  250. inspect_ai/_view/www/src/workspace/sidebar/Sidebar.tsx +85 -0
  251. inspect_ai/_view/www/src/workspace/sidebar/SidebarLogEntry.module.css +29 -0
  252. inspect_ai/_view/www/src/workspace/sidebar/SidebarLogEntry.tsx +95 -0
  253. inspect_ai/_view/www/src/workspace/sidebar/SidebarScoreView.module.css +23 -0
  254. inspect_ai/_view/www/src/workspace/sidebar/SidebarScoreView.tsx +41 -0
  255. inspect_ai/_view/www/src/workspace/sidebar/SidebarScoresView.module.css +35 -0
  256. inspect_ai/_view/www/src/workspace/sidebar/SidebarScoresView.tsx +61 -0
  257. inspect_ai/_view/www/src/workspace/tabs/InfoTab.tsx +80 -0
  258. inspect_ai/_view/www/src/workspace/tabs/JsonTab.module.css +5 -0
  259. inspect_ai/_view/www/src/workspace/tabs/JsonTab.tsx +46 -0
  260. inspect_ai/_view/www/src/workspace/tabs/SamplesTab.tsx +204 -0
  261. inspect_ai/_view/www/src/workspace/tabs/grouping.ts +195 -0
  262. inspect_ai/_view/www/src/workspace/tabs/types.ts +19 -0
  263. inspect_ai/_view/www/src/workspace/types.ts +10 -0
  264. inspect_ai/_view/www/tsconfig.json +23 -9
  265. inspect_ai/_view/www/vite.config.js +8 -17
  266. inspect_ai/_view/www/yarn.lock +627 -556
  267. inspect_ai/dataset/_dataset.py +36 -0
  268. inspect_ai/dataset/_sources/csv.py +8 -0
  269. inspect_ai/dataset/_sources/file.py +4 -0
  270. inspect_ai/dataset/_sources/hf.py +11 -1
  271. inspect_ai/dataset/_sources/json.py +8 -0
  272. inspect_ai/log/_log.py +3 -6
  273. inspect_ai/log/_message.py +1 -1
  274. inspect_ai/log/_recorders/json.py +5 -7
  275. inspect_ai/model/_call_tools.py +2 -1
  276. inspect_ai/model/_providers/anthropic.py +3 -3
  277. inspect_ai/model/_providers/openai_o1.py +3 -5
  278. inspect_ai/model/_providers/openrouter.py +86 -0
  279. inspect_ai/model/_providers/providers.py +11 -0
  280. inspect_ai/scorer/_answer.py +7 -7
  281. inspect_ai/scorer/_classification.py +34 -18
  282. inspect_ai/scorer/_common.py +2 -8
  283. inspect_ai/solver/_multiple_choice.py +24 -9
  284. inspect_ai/tool/__init__.py +2 -0
  285. inspect_ai/tool/{beta → _tools}/_computer/_computer.py +2 -5
  286. inspect_ai/tool/_tools/_computer/_resources/tool/__init__.py +0 -0
  287. inspect_ai/tool/{beta → _tools}/_computer/_resources/tool/_x11_client.py +1 -1
  288. inspect_ai/tool/_tools/_computer/_resources/tool/requirements.txt +0 -0
  289. inspect_ai/tool/_tools/_execute.py +8 -2
  290. inspect_ai/tool/beta.py +3 -0
  291. inspect_ai/util/_sandbox/docker/docker.py +32 -85
  292. inspect_ai/util/_sandbox/self_check.py +124 -16
  293. {inspect_ai-0.3.62.dist-info → inspect_ai-0.3.63.dist-info}/METADATA +2 -1
  294. inspect_ai-0.3.63.dist-info/RECORD +618 -0
  295. inspect_ai/_view/www/src/Register.mjs +0 -3
  296. inspect_ai/_view/www/src/Types.mjs +0 -38
  297. inspect_ai/_view/www/src/appearance/Colors.mjs +0 -27
  298. inspect_ai/_view/www/src/appearance/Fonts.mjs +0 -66
  299. inspect_ai/_view/www/src/appearance/Icons.mjs +0 -240
  300. inspect_ai/_view/www/src/components/AnsiDisplay.mjs +0 -184
  301. inspect_ai/_view/www/src/components/AppErrorBoundary.mjs +0 -34
  302. inspect_ai/_view/www/src/components/AsciiCinemaPlayer.mjs +0 -74
  303. inspect_ai/_view/www/src/components/Card.mjs +0 -126
  304. inspect_ai/_view/www/src/components/ChatView.mjs +0 -441
  305. inspect_ai/_view/www/src/components/CopyButton.mjs +0 -48
  306. inspect_ai/_view/www/src/components/Dialog.mjs +0 -61
  307. inspect_ai/_view/www/src/components/DownloadButton.mjs +0 -15
  308. inspect_ai/_view/www/src/components/DownloadPanel.mjs +0 -29
  309. inspect_ai/_view/www/src/components/EmptyPanel.mjs +0 -23
  310. inspect_ai/_view/www/src/components/ErrorPanel.mjs +0 -66
  311. inspect_ai/_view/www/src/components/ExpandablePanel.mjs +0 -136
  312. inspect_ai/_view/www/src/components/FindBand.mjs +0 -157
  313. inspect_ai/_view/www/src/components/HumanBaselineView.mjs +0 -168
  314. inspect_ai/_view/www/src/components/JsonPanel.mjs +0 -61
  315. inspect_ai/_view/www/src/components/LabeledValue.mjs +0 -32
  316. inspect_ai/_view/www/src/components/LargeModal.mjs +0 -190
  317. inspect_ai/_view/www/src/components/LightboxCarousel.mjs +0 -217
  318. inspect_ai/_view/www/src/components/MarkdownDiv.mjs +0 -118
  319. inspect_ai/_view/www/src/components/MessageBand.mjs +0 -48
  320. inspect_ai/_view/www/src/components/MessageContent.mjs +0 -111
  321. inspect_ai/_view/www/src/components/MetaDataGrid.mjs +0 -92
  322. inspect_ai/_view/www/src/components/MetaDataView.mjs +0 -109
  323. inspect_ai/_view/www/src/components/MorePopOver.mjs +0 -50
  324. inspect_ai/_view/www/src/components/NavPills.mjs +0 -63
  325. inspect_ai/_view/www/src/components/ProgressBar.mjs +0 -51
  326. inspect_ai/_view/www/src/components/RenderedContent/ChatMessageRenderer.mjs +0 -54
  327. inspect_ai/_view/www/src/components/RenderedContent/Types.mjs +0 -19
  328. inspect_ai/_view/www/src/components/TabSet.mjs +0 -184
  329. inspect_ai/_view/www/src/components/ToolButton.mjs +0 -16
  330. inspect_ai/_view/www/src/components/Tools.mjs +0 -376
  331. inspect_ai/_view/www/src/components/VirtualList.mjs +0 -280
  332. inspect_ai/_view/www/src/components/ansi-output.js +0 -932
  333. inspect_ai/_view/www/src/json/JsonTab.mjs +0 -48
  334. inspect_ai/_view/www/src/log-reader/Log-Reader.mjs +0 -25
  335. inspect_ai/_view/www/src/log-reader/Native-Log-Reader.mjs +0 -13
  336. inspect_ai/_view/www/src/log-reader/Open-AI-Log-Reader.mjs +0 -263
  337. inspect_ai/_view/www/src/navbar/Navbar.mjs +0 -418
  338. inspect_ai/_view/www/src/navbar/SecondaryBar.mjs +0 -175
  339. inspect_ai/_view/www/src/plan/PlanCard.mjs +0 -418
  340. inspect_ai/_view/www/src/samples/SampleDialog.mjs +0 -123
  341. inspect_ai/_view/www/src/samples/SampleDisplay.mjs +0 -516
  342. inspect_ai/_view/www/src/samples/SampleError.mjs +0 -99
  343. inspect_ai/_view/www/src/samples/SampleList.mjs +0 -427
  344. inspect_ai/_view/www/src/samples/SampleScoreView.mjs +0 -172
  345. inspect_ai/_view/www/src/samples/SampleScores.mjs +0 -34
  346. inspect_ai/_view/www/src/samples/SampleTranscript.mjs +0 -20
  347. inspect_ai/_view/www/src/samples/SamplesDescriptor.mjs +0 -771
  348. inspect_ai/_view/www/src/samples/SamplesTab.mjs +0 -399
  349. inspect_ai/_view/www/src/samples/SamplesTools.mjs +0 -64
  350. inspect_ai/_view/www/src/samples/tools/EpochFilter.mjs +0 -38
  351. inspect_ai/_view/www/src/samples/tools/SampleFilter.mjs +0 -756
  352. inspect_ai/_view/www/src/samples/tools/SelectScorer.mjs +0 -141
  353. inspect_ai/_view/www/src/samples/tools/SortFilter.mjs +0 -151
  354. inspect_ai/_view/www/src/samples/transcript/ApprovalEventView.mjs +0 -71
  355. inspect_ai/_view/www/src/samples/transcript/ErrorEventView.mjs +0 -44
  356. inspect_ai/_view/www/src/samples/transcript/EventPanel.mjs +0 -271
  357. inspect_ai/_view/www/src/samples/transcript/EventRow.mjs +0 -46
  358. inspect_ai/_view/www/src/samples/transcript/EventSection.mjs +0 -33
  359. inspect_ai/_view/www/src/samples/transcript/InfoEventView.mjs +0 -59
  360. inspect_ai/_view/www/src/samples/transcript/InputEventView.mjs +0 -44
  361. inspect_ai/_view/www/src/samples/transcript/LoggerEventView.mjs +0 -32
  362. inspect_ai/_view/www/src/samples/transcript/ModelEventView.mjs +0 -216
  363. inspect_ai/_view/www/src/samples/transcript/SampleInitEventView.mjs +0 -107
  364. inspect_ai/_view/www/src/samples/transcript/SampleLimitEventView.mjs +0 -74
  365. inspect_ai/_view/www/src/samples/transcript/ScoreEventView.mjs +0 -100
  366. inspect_ai/_view/www/src/samples/transcript/StepEventView.mjs +0 -187
  367. inspect_ai/_view/www/src/samples/transcript/SubtaskEventView.mjs +0 -133
  368. inspect_ai/_view/www/src/samples/transcript/ToolEventView.mjs +0 -88
  369. inspect_ai/_view/www/src/samples/transcript/TranscriptView.mjs +0 -459
  370. inspect_ai/_view/www/src/samples/transcript/Types.mjs +0 -44
  371. inspect_ai/_view/www/src/samples/transcript/state/StateDiffView.mjs +0 -53
  372. inspect_ai/_view/www/src/samples/transcript/state/StateEventRenderers.mjs +0 -254
  373. inspect_ai/_view/www/src/sidebar/Sidebar.mjs +0 -418
  374. inspect_ai/_view/www/src/usage/ModelTokenTable.mjs +0 -72
  375. inspect_ai/_view/www/src/usage/UsageCard.mjs +0 -159
  376. inspect_ai/_view/www/src/utils/Format.mjs +0 -260
  377. inspect_ai/_view/www/src/utils/Git.mjs +0 -12
  378. inspect_ai/_view/www/src/utils/Html.mjs +0 -21
  379. inspect_ai/_view/www/src/utils/attachments.mjs +0 -31
  380. inspect_ai/_view/www/src/utils/debugging.mjs +0 -23
  381. inspect_ai/_view/www/src/utils/http.mjs +0 -18
  382. inspect_ai/_view/www/src/utils/queue.mjs +0 -67
  383. inspect_ai/_view/www/src/utils/sync.mjs +0 -101
  384. inspect_ai/_view/www/src/workspace/TaskErrorPanel.mjs +0 -17
  385. inspect_ai/_view/www/src/workspace/WorkSpace.mjs +0 -516
  386. inspect_ai/tool/beta/__init__.py +0 -5
  387. inspect_ai-0.3.62.dist-info/RECORD +0 -481
  388. /inspect_ai/{tool/beta/_computer/_resources/tool/__init__.py → _view/www/src/components/MorePopOver.css} +0 -0
  389. /inspect_ai/_view/www/src/{constants.mjs → constants.ts} +0 -0
  390. /inspect_ai/{tool/beta/_computer/_resources/tool/requirements.txt → _view/www/src/workspace/tabs/InfoTab.module.css} +0 -0
  391. /inspect_ai/tool/{beta → _tools}/_computer/__init__.py +0 -0
  392. /inspect_ai/tool/{beta → _tools}/_computer/_common.py +0 -0
  393. /inspect_ai/tool/{beta → _tools}/_computer/_computer_split.py +0 -0
  394. /inspect_ai/tool/{beta → _tools}/_computer/_resources/Dockerfile +0 -0
  395. /inspect_ai/tool/{beta → _tools}/_computer/_resources/README.md +0 -0
  396. /inspect_ai/tool/{beta → _tools}/_computer/_resources/entrypoint/entrypoint.sh +0 -0
  397. /inspect_ai/tool/{beta → _tools}/_computer/_resources/entrypoint/novnc_startup.sh +0 -0
  398. /inspect_ai/tool/{beta → _tools}/_computer/_resources/entrypoint/x11vnc_startup.sh +0 -0
  399. /inspect_ai/tool/{beta → _tools}/_computer/_resources/entrypoint/xfce_startup.sh +0 -0
  400. /inspect_ai/tool/{beta → _tools}/_computer/_resources/entrypoint/xvfb_startup.sh +0 -0
  401. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/.config/Code/User/globalStorage/state.vscdb +0 -0
  402. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/.config/Code/User/settings.json +0 -0
  403. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/.config/xfce4/xfconf/xfce-perchannel-xml/xfce4-panel.xml +0 -0
  404. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/.config/xfce4/xfconf/xfce-perchannel-xml/xfce4-screensaver.xml +0 -0
  405. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/Desktop/Firefox Web Browser.desktop +0 -0
  406. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/Desktop/Terminal.desktop +0 -0
  407. /inspect_ai/tool/{beta → _tools}/_computer/_resources/image_home_dir/Desktop/Visual Studio Code.desktop +0 -0
  408. /inspect_ai/tool/{beta → _tools}/_computer/_resources/tool/_logger.py +0 -0
  409. /inspect_ai/tool/{beta → _tools}/_computer/_resources/tool/_run.py +0 -0
  410. /inspect_ai/tool/{beta → _tools}/_computer/_resources/tool/_tool_result.py +0 -0
  411. /inspect_ai/tool/{beta → _tools}/_computer/_resources/tool/computer_tool.py +0 -0
  412. {inspect_ai-0.3.62.dist-info → inspect_ai-0.3.63.dist-info}/LICENSE +0 -0
  413. {inspect_ai-0.3.62.dist-info → inspect_ai-0.3.63.dist-info}/WHEEL +0 -0
  414. {inspect_ai-0.3.62.dist-info → inspect_ai-0.3.63.dist-info}/entry_points.txt +0 -0
  415. {inspect_ai-0.3.62.dist-info → inspect_ai-0.3.63.dist-info}/top_level.txt +0 -0
@@ -0,0 +1,326 @@
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
+ label="Print"
81
+ icon={ApplicationIcons.copy}
82
+ onClick={() => {
83
+ printSample(id, targetId);
84
+ }}
85
+ />,
86
+ );
87
+ }
88
+
89
+ return (
90
+ <Fragment>
91
+ <SampleSummaryView
92
+ parent_id={id}
93
+ sample={sample}
94
+ sampleDescriptor={sampleDescriptor}
95
+ />
96
+ <TabSet
97
+ id={tabsetId}
98
+ tabControlsClassName={clsx("text-size-base")}
99
+ tabPanelsClassName={clsx(styles.tabPanel)}
100
+ tools={tools}
101
+ >
102
+ {sample.events && sample.events.length > 0 ? (
103
+ <TabPanel
104
+ id={kSampleTranscriptTabId}
105
+ className="sample-tab"
106
+ title="Transcript"
107
+ onSelected={onSelectedTab}
108
+ selected={
109
+ selectedTab === kSampleTranscriptTabId ||
110
+ selectedTab === undefined
111
+ }
112
+ scrollable={false}
113
+ >
114
+ <SampleTranscript
115
+ key={`${baseId}-transcript-display-${id}`}
116
+ id={`${baseId}-transcript-display-${id}`}
117
+ evalEvents={sample.events}
118
+ scrollRef={scrollRef}
119
+ />
120
+ </TabPanel>
121
+ ) : null}
122
+ <TabPanel
123
+ id={kSampleMessagesTabId}
124
+ className={clsx("sample-tab", styles.fullWidth)}
125
+ title="Messages"
126
+ onSelected={onSelectedTab}
127
+ selected={selectedTab === kSampleMessagesTabId}
128
+ scrollable={false}
129
+ >
130
+ <ChatViewVirtualList
131
+ key={`${baseId}-chat-${id}`}
132
+ id={`${baseId}-chat-${id}`}
133
+ messages={sample.messages}
134
+ indented={true}
135
+ scrollRef={scrollRef}
136
+ toolCallStyle="complete"
137
+ />
138
+ </TabPanel>
139
+ {scorerNames.length === 1 ? (
140
+ <TabPanel
141
+ id={kSampleScoringTabId}
142
+ className="sample-tab"
143
+ title="Scoring"
144
+ onSelected={onSelectedTab}
145
+ selected={selectedTab === kSampleScoringTabId}
146
+ >
147
+ <SampleScoreView
148
+ sample={sample}
149
+ sampleDescriptor={sampleDescriptor}
150
+ scorer={scorerNames[0]}
151
+ />
152
+ </TabPanel>
153
+ ) : (
154
+ <>
155
+ {Object.keys(sample.scores || {}).map((scorer) => {
156
+ const tabId = `score-${scorer}`;
157
+ return (
158
+ <TabPanel
159
+ id={tabId}
160
+ className="sample-tab"
161
+ title={scorer}
162
+ onSelected={onSelectedTab}
163
+ selected={selectedTab === tabId}
164
+ >
165
+ <SampleScoreView
166
+ sample={sample}
167
+ sampleDescriptor={sampleDescriptor}
168
+ scorer={scorer}
169
+ />
170
+ </TabPanel>
171
+ );
172
+ })}
173
+ </>
174
+ )}
175
+ {sampleMetadatas.length > 0 ? (
176
+ <TabPanel
177
+ id={kSampleMetdataTabId}
178
+ className="sample-tab"
179
+ title="Metadata"
180
+ onSelected={onSelectedTab}
181
+ selected={selectedTab === kSampleMetdataTabId}
182
+ >
183
+ <div className={clsx(styles.metadataPanel)}>{sampleMetadatas}</div>
184
+ </TabPanel>
185
+ ) : null}
186
+ {sample.error ? (
187
+ <TabPanel
188
+ id={kSampleErrorTabId}
189
+ className="sample-tab"
190
+ title="Error"
191
+ onSelected={onSelectedTab}
192
+ selected={selectedTab === kSampleErrorTabId}
193
+ >
194
+ <div className={clsx(styles.padded)}>
195
+ <ANSIDisplay
196
+ output={sample.error.traceback_ansi}
197
+ className={clsx("text-size-small", styles.ansi)}
198
+ />
199
+ </div>
200
+ </TabPanel>
201
+ ) : null}
202
+ {sample.messages.length < 100 ? (
203
+ <TabPanel
204
+ id={kSampleJsonTabId}
205
+ className={"sample-tab"}
206
+ title="JSON"
207
+ onSelected={onSelectedTab}
208
+ selected={selectedTab === kSampleJsonTabId}
209
+ >
210
+ <div className={clsx(styles.padded, styles.fullWidth)}>
211
+ <JSONPanel data={sample} simple={true} />
212
+ </div>
213
+ </TabPanel>
214
+ ) : null}
215
+ </TabSet>
216
+ </Fragment>
217
+ );
218
+ };
219
+
220
+ const metadataViewsForSample = (_id: string, sample: EvalSample) => {
221
+ const sampleMetadatas = [];
222
+ if (sample.model_usage && Object.keys(sample.model_usage).length > 0) {
223
+ sampleMetadatas.push(
224
+ <Card>
225
+ <CardHeader label="Usage" />
226
+ <CardBody>
227
+ <ModelTokenTable
228
+ model_usage={sample.model_usage}
229
+ className={clsx(styles.noTop)}
230
+ />
231
+ </CardBody>
232
+ </Card>,
233
+ );
234
+ }
235
+
236
+ if (Object.keys(sample?.metadata).length > 0) {
237
+ sampleMetadatas.push(
238
+ <Card>
239
+ <CardHeader label="Metadata" />
240
+ <CardBody>
241
+ <MetaDataView
242
+ id="task-sample-metadata-${id}"
243
+ entries={sample?.metadata as Record<string, unknown>}
244
+ className={clsx("tab-pane", styles.noTop)}
245
+ />
246
+ </CardBody>
247
+ </Card>,
248
+ );
249
+ }
250
+
251
+ if (Object.keys(sample?.store).length > 0) {
252
+ sampleMetadatas.push(
253
+ <Card>
254
+ <CardHeader label="Store" />
255
+ <CardBody>
256
+ <MetaDataView
257
+ id="task-sample-store-${id}"
258
+ entries={sample?.store as Record<string, unknown>}
259
+ className={clsx("tab-pane", styles.noTop)}
260
+ />
261
+ </CardBody>
262
+ </Card>,
263
+ );
264
+ }
265
+
266
+ return sampleMetadatas;
267
+ };
268
+ const printSample = (id: string, targetId: string) => {
269
+ // The active tab
270
+ const targetTabEl = document.querySelector(
271
+ `#${escapeSelector(targetId)} .sample-tab.tab-pane.show.active`,
272
+ );
273
+ if (targetTabEl) {
274
+ // The target element
275
+ const targetEl = targetTabEl.firstElementChild;
276
+ if (targetEl) {
277
+ // Get the sample heading to include
278
+ const headingId = `sample-heading-${id}`;
279
+ const headingEl = document.getElementById(headingId);
280
+
281
+ // Print the document
282
+ const headingHtml = printHeadingHtml();
283
+ const css = `
284
+ html { font-size: 9pt }
285
+ /* Allow content to break anywhere without any forced page breaks */
286
+ * {
287
+ break-inside: auto; /* Let elements break anywhere */
288
+ page-break-inside: auto; /* Legacy support */
289
+ break-before: auto;
290
+ page-break-before: auto;
291
+ break-after: auto;
292
+ page-break-after: auto;
293
+ }
294
+ /* Specifically disable all page breaks for divs */
295
+ div {
296
+ break-inside: auto;
297
+ page-break-inside: auto;
298
+ }
299
+ body > .transcript-step {
300
+ break-inside: avoid;
301
+ }
302
+ body{
303
+ -webkit-print-color-adjust:exact !important;
304
+ print-color-adjust:exact !important;
305
+ }
306
+ /* Allow preformatted text and code blocks to break across pages */
307
+ pre, code {
308
+ white-space: pre-wrap; /* Wrap long lines instead of keeping them on one line */
309
+ overflow-wrap: break-word; /* Ensure long words are broken to fit within the page */
310
+ break-inside: auto; /* Allow page breaks inside the element */
311
+ page-break-inside: auto; /* Older equivalent */
312
+ }
313
+
314
+ /* Additional control for long lines within code/preformatted blocks */
315
+ pre {
316
+ word-wrap: break-word; /* Break long words if needed */
317
+ }
318
+
319
+ `;
320
+ printHtml(
321
+ [headingHtml, headingEl?.outerHTML, targetEl.innerHTML].join("\n"),
322
+ css,
323
+ );
324
+ }
325
+ }
326
+ };
@@ -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 +1,175 @@
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) => {
147
+ return (
148
+ <div
149
+ className={clsx(
150
+ "text-style-label",
151
+ "text-style-secondary",
152
+ "text-size-base",
153
+ col.center ? styles.centerLabel : undefined,
154
+ )}
155
+ >
156
+ {col.label}
157
+ </div>
158
+ );
159
+ })}
160
+ {columns.map((col) => {
161
+ return (
162
+ <div
163
+ className={clsx(
164
+ styles.wrap,
165
+ col.clamp ? "three-line-clamp" : undefined,
166
+ col.center ? styles.centerLabel : undefined,
167
+ )}
168
+ >
169
+ {col.value}
170
+ </div>
171
+ );
172
+ })}
173
+ </div>
174
+ );
175
+ };
@@ -0,0 +1,60 @@
1
+ import { ScoreFilter, ScoreLabel } from "../types";
2
+ import { SamplesDescriptor } from "./descriptor/samplesDescriptor";
3
+ import { EpochFilter } from "./sample-tools/EpochFilter";
4
+ import { SampleFilter } from "./sample-tools/sample-filter/SampleFilter";
5
+ import { SelectScorer } from "./sample-tools/SelectScorer";
6
+ import { SortFilter } from "./sample-tools/SortFilter";
7
+
8
+ interface SampleToolsProps {
9
+ epoch: string;
10
+ setEpoch: (epoch: string) => void;
11
+ epochs: number;
12
+ scoreFilter: ScoreFilter;
13
+ setScoreFilter: (filter: ScoreFilter) => void;
14
+ sort: string;
15
+ setSort: (sort: string) => void;
16
+ score?: ScoreLabel;
17
+ setScore: (score: ScoreLabel) => void;
18
+ scores: ScoreLabel[];
19
+ sampleDescriptor: SamplesDescriptor;
20
+ }
21
+
22
+ export const SampleTools: React.FC<SampleToolsProps> = ({
23
+ epoch,
24
+ setEpoch,
25
+ epochs,
26
+ scoreFilter,
27
+ setScoreFilter,
28
+ sort,
29
+ setSort,
30
+ score,
31
+ setScore,
32
+ scores,
33
+ sampleDescriptor,
34
+ }) => {
35
+ const tools = [];
36
+
37
+ tools.push(
38
+ <SampleFilter
39
+ evalDescriptor={sampleDescriptor.evalDescriptor}
40
+ scoreFilter={scoreFilter}
41
+ setScoreFilter={setScoreFilter}
42
+ />,
43
+ );
44
+
45
+ if (scores.length > 1) {
46
+ tools.push(
47
+ <SelectScorer scores={scores} score={score} setScore={setScore} />,
48
+ );
49
+ }
50
+
51
+ if (epochs > 1) {
52
+ tools.push(
53
+ <EpochFilter epoch={epoch} setEpoch={setEpoch} epochs={epochs} />,
54
+ );
55
+ }
56
+
57
+ tools.push(<SortFilter sort={sort} setSort={setSort} epochs={epochs} />);
58
+
59
+ return tools;
60
+ };
@@ -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>
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
+ };