inspect-ai 0.3.90__py3-none-any.whl → 0.3.92__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.
- inspect_ai/_cli/common.py +13 -0
- inspect_ai/_cli/eval.py +44 -0
- inspect_ai/_display/textual/widgets/samples.py +49 -4
- inspect_ai/_display/textual/widgets/vscode.py +4 -2
- inspect_ai/_eval/eval.py +41 -28
- inspect_ai/_eval/evalset.py +4 -0
- inspect_ai/_eval/loader.py +4 -5
- inspect_ai/_eval/registry.py +1 -1
- inspect_ai/_eval/run.py +6 -3
- inspect_ai/_eval/task/log.py +6 -0
- inspect_ai/_eval/task/run.py +108 -41
- inspect_ai/_eval/task/sandbox.py +19 -5
- inspect_ai/_util/_async.py +1 -1
- inspect_ai/_util/constants.py +1 -0
- inspect_ai/_util/environ.py +32 -0
- inspect_ai/_util/file.py +8 -1
- inspect_ai/_util/httpx.py +105 -22
- inspect_ai/_util/registry.py +83 -9
- inspect_ai/_util/text.py +81 -17
- inspect_ai/_util/transcript.py +9 -6
- inspect_ai/_util/vscode.py +7 -2
- inspect_ai/_view/schema.py +1 -1
- inspect_ai/_view/www/babel.config.js +11 -0
- inspect_ai/_view/www/dist/assets/index.css +3640 -3563
- inspect_ai/_view/www/dist/assets/index.js +59204 -52519
- inspect_ai/_view/www/eslint.config.mjs +10 -1
- inspect_ai/_view/www/jest.config.mjs +21 -0
- inspect_ai/_view/www/log-schema.json +111 -2
- inspect_ai/_view/www/package.json +19 -5
- inspect_ai/_view/www/src/{types → @types}/log.d.ts +95 -32
- inspect_ai/_view/www/{App.css → src/app/App.css} +22 -14
- inspect_ai/_view/www/src/app/App.tsx +168 -0
- inspect_ai/_view/www/src/{AppErrorBoundary.tsx → app/AppErrorBoundary.tsx} +1 -1
- inspect_ai/_view/www/src/{appearance → app/appearance}/icons.ts +1 -0
- inspect_ai/_view/www/src/{metadata → app/content}/RenderedContent.tsx +5 -5
- inspect_ai/_view/www/src/{workspace/WorkSpaceView.tsx → app/log-view/LogView.tsx} +59 -40
- inspect_ai/_view/www/src/app/log-view/LogViewContainer.tsx +159 -0
- inspect_ai/_view/www/src/app/log-view/LogViewLayout.tsx +109 -0
- inspect_ai/_view/www/src/{workspace → app/log-view}/error/TaskErrorPanel.tsx +3 -3
- inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/ModelRolesView.tsx +1 -1
- inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/Navbar.tsx +4 -4
- inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/PrimaryBar.tsx +8 -8
- inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/ResultsPanel.tsx +6 -6
- inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/RunningStatusPanel.tsx +1 -1
- inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/ScoreGrid.tsx +1 -1
- inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/SecondaryBar.tsx +8 -8
- inspect_ai/_view/www/src/{workspace → app/log-view}/tabs/InfoTab.tsx +35 -6
- inspect_ai/_view/www/src/app/log-view/tabs/JsonTab.tsx +136 -0
- inspect_ai/_view/www/src/{workspace → app/log-view}/tabs/SamplesTab.tsx +82 -73
- inspect_ai/_view/www/src/{workspace → app/log-view}/tabs/grouping.ts +3 -3
- inspect_ai/_view/www/src/{workspace → app/log-view}/tabs/types.ts +1 -1
- inspect_ai/_view/www/src/{plan → app/plan}/DatasetDetailView.tsx +2 -2
- inspect_ai/_view/www/src/{plan → app/plan}/DetailStep.tsx +1 -1
- inspect_ai/_view/www/src/{plan → app/plan}/ModelCard.tsx +4 -4
- inspect_ai/_view/www/src/{plan → app/plan}/PlanCard.tsx +2 -2
- inspect_ai/_view/www/src/{plan → app/plan}/PlanDetailView.tsx +5 -5
- inspect_ai/_view/www/src/{plan → app/plan}/SolverDetailView.tsx +1 -1
- inspect_ai/_view/www/src/app/routing/AppRouter.tsx +58 -0
- inspect_ai/_view/www/src/app/routing/navigationHooks.ts +182 -0
- inspect_ai/_view/www/src/app/routing/url.ts +43 -0
- inspect_ai/_view/www/src/{samples → app/samples}/InlineSampleDisplay.tsx +11 -27
- inspect_ai/_view/www/src/{samples → app/samples}/SampleDialog.tsx +36 -40
- inspect_ai/_view/www/src/{samples → app/samples}/SampleDisplay.module.css +4 -0
- inspect_ai/_view/www/src/{samples → app/samples}/SampleDisplay.tsx +116 -49
- inspect_ai/_view/www/src/{samples → app/samples}/SampleSummaryView.module.css +1 -1
- inspect_ai/_view/www/src/{samples → app/samples}/SampleSummaryView.tsx +29 -26
- inspect_ai/_view/www/src/{samples → app/samples}/SamplesTools.tsx +3 -3
- inspect_ai/_view/www/src/{samples → app/samples}/chat/ChatMessage.module.css +5 -2
- inspect_ai/_view/www/src/{samples → app/samples}/chat/ChatMessage.tsx +12 -4
- inspect_ai/_view/www/src/{samples → app/samples}/chat/ChatMessageRenderer.tsx +3 -3
- inspect_ai/_view/www/src/{samples → app/samples}/chat/ChatMessageRow.tsx +6 -1
- inspect_ai/_view/www/src/{samples → app/samples}/chat/ChatView.tsx +4 -2
- inspect_ai/_view/www/src/{samples → app/samples}/chat/ChatViewVirtualList.tsx +5 -3
- inspect_ai/_view/www/src/app/samples/chat/MessageContent.module.css +12 -0
- inspect_ai/_view/www/src/{samples → app/samples}/chat/MessageContent.tsx +11 -10
- inspect_ai/_view/www/src/app/samples/chat/MessageContents.module.css +7 -0
- inspect_ai/_view/www/src/{samples → app/samples}/chat/MessageContents.tsx +14 -8
- inspect_ai/_view/www/src/{samples → app/samples}/chat/messages.ts +2 -2
- inspect_ai/_view/www/src/app/samples/chat/tools/ToolCallView.module.css +7 -0
- inspect_ai/_view/www/src/{samples → app/samples}/chat/tools/ToolCallView.tsx +26 -27
- inspect_ai/_view/www/src/app/samples/chat/tools/ToolInput.module.css +19 -0
- inspect_ai/_view/www/src/{samples → app/samples}/chat/tools/ToolInput.tsx +3 -3
- inspect_ai/_view/www/src/{samples → app/samples}/chat/tools/ToolOutput.module.css +1 -0
- inspect_ai/_view/www/src/{samples → app/samples}/chat/tools/ToolOutput.tsx +1 -1
- inspect_ai/_view/www/src/{samples → app/samples}/chat/tools/ToolTitle.module.css +4 -0
- inspect_ai/_view/www/src/{samples → app/samples}/chat/tools/ToolTitle.tsx +2 -2
- inspect_ai/_view/www/src/{samples → app/samples}/chat/tools/tool.ts +1 -1
- inspect_ai/_view/www/src/app/samples/chat/types.ts +1 -0
- inspect_ai/_view/www/src/{samples → app/samples}/descriptor/samplesDescriptor.tsx +38 -15
- inspect_ai/_view/www/src/{samples → app/samples}/descriptor/score/BooleanScoreDescriptor.tsx +1 -1
- inspect_ai/_view/www/src/{samples → app/samples}/descriptor/score/CategoricalScoreDescriptor.tsx +2 -2
- inspect_ai/_view/www/src/{samples → app/samples}/descriptor/score/NumericScoreDescriptor.tsx +3 -3
- inspect_ai/_view/www/src/{samples → app/samples}/descriptor/score/ObjectScoreDescriptor.tsx +4 -4
- inspect_ai/_view/www/src/{samples → app/samples}/descriptor/score/OtherScoreDescriptor.tsx +2 -2
- inspect_ai/_view/www/src/{samples → app/samples}/descriptor/score/PassFailScoreDescriptor.tsx +2 -2
- inspect_ai/_view/www/src/{samples → app/samples}/descriptor/score/ScoreDescriptor.tsx +1 -1
- inspect_ai/_view/www/src/{samples → app/samples}/descriptor/types.ts +4 -3
- inspect_ai/_view/www/src/{samples → app/samples}/error/SampleErrorView.module.css +2 -1
- inspect_ai/_view/www/src/{samples → app/samples}/list/SampleHeader.tsx +3 -0
- inspect_ai/_view/www/src/{samples → app/samples}/list/SampleList.tsx +47 -33
- inspect_ai/_view/www/src/{samples → app/samples}/list/SampleRow.module.css +16 -0
- inspect_ai/_view/www/src/{samples → app/samples}/list/SampleRow.tsx +47 -20
- inspect_ai/_view/www/src/{samples → app/samples}/sample-tools/SelectScorer.tsx +1 -1
- inspect_ai/_view/www/src/{samples → app/samples}/sample-tools/SortFilter.tsx +4 -4
- inspect_ai/_view/www/src/{samples → app/samples}/sample-tools/filters.ts +8 -6
- inspect_ai/_view/www/src/{samples → app/samples}/sample-tools/sample-filter/SampleFilter.tsx +4 -3
- inspect_ai/_view/www/src/{samples → app/samples}/sample-tools/sample-filter/completions.ts +1 -1
- inspect_ai/_view/www/src/{samples → app/samples}/sample-tools/sample-filter/language.ts +1 -0
- inspect_ai/_view/www/src/{samples → app/samples}/sampleDataAdapter.ts +3 -3
- inspect_ai/_view/www/src/{samples → app/samples}/sampleLimit.ts +1 -1
- inspect_ai/_view/www/src/{samples → app/samples}/scores/SampleScores.tsx +1 -1
- inspect_ai/_view/www/src/{samples → app/samples}/scores/SampleScoresGrid.tsx +12 -11
- inspect_ai/_view/www/src/{samples → app/samples}/scores/SampleScoresView.tsx +6 -6
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/ApprovalEventView.tsx +1 -1
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/ErrorEventView.tsx +3 -3
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/InfoEventView.tsx +4 -4
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/InputEventView.tsx +3 -3
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/LoggerEventView.tsx +3 -3
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/ModelEventView.module.css +13 -7
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/ModelEventView.tsx +49 -21
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/SampleInitEventView.tsx +11 -9
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/SampleLimitEventView.tsx +1 -1
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/SandboxEventView.tsx +8 -6
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/ScoreEventView.tsx +4 -4
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/StepEventView.tsx +11 -3
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/SubtaskEventView.tsx +2 -2
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/ToolEventView.tsx +2 -2
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/TranscriptView.module.css +8 -7
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/TranscriptView.tsx +32 -114
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/TranscriptVirtualListComponent.module.css +6 -5
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/TranscriptVirtualListComponent.tsx +14 -2
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventPanel.tsx +2 -2
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventTimingPanel.tsx +1 -1
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/utils.ts +1 -1
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/state/StateEventRenderers.tsx +23 -21
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/state/StateEventRenders.module.css +7 -0
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/state/StateEventView.tsx +2 -2
- inspect_ai/_view/www/src/app/samples/transcript/transform/fixups.ts +142 -0
- inspect_ai/_view/www/src/app/samples/transcript/transform/treeify.ts +39 -0
- inspect_ai/_view/www/src/{samples → app/samples}/transcript/types.ts +1 -1
- inspect_ai/_view/www/src/{workspace → app}/sidebar/EvalStatus.tsx +1 -1
- inspect_ai/_view/www/src/app/sidebar/LogDirectoryTitleView.module.css +16 -0
- inspect_ai/_view/www/src/app/sidebar/LogDirectoryTitleView.tsx +70 -0
- inspect_ai/_view/www/src/{workspace → app}/sidebar/Sidebar.module.css +8 -0
- inspect_ai/_view/www/src/{workspace → app}/sidebar/Sidebar.tsx +35 -17
- inspect_ai/_view/www/src/{workspace → app}/sidebar/SidebarLogEntry.tsx +1 -1
- inspect_ai/_view/www/src/{workspace → app}/sidebar/SidebarScoreView.tsx +2 -2
- inspect_ai/_view/www/src/{workspace → app}/sidebar/SidebarScoresView.tsx +2 -2
- inspect_ai/_view/www/src/{types.ts → app/types.ts} +18 -11
- inspect_ai/_view/www/src/{usage → app/usage}/ModelTokenTable.tsx +1 -1
- inspect_ai/_view/www/src/{usage → app/usage}/ModelUsagePanel.tsx +2 -2
- inspect_ai/_view/www/src/{usage → app/usage}/TokenTable.tsx +1 -1
- inspect_ai/_view/www/src/{usage → app/usage}/UsageCard.tsx +6 -6
- inspect_ai/_view/www/src/{api → client/api}/api-browser.ts +2 -2
- inspect_ai/_view/www/src/{api → client/api}/api-http.ts +3 -3
- inspect_ai/_view/www/src/{api → client/api}/api-vscode.ts +2 -2
- inspect_ai/_view/www/src/{api → client/api}/client-api.ts +6 -5
- inspect_ai/_view/www/src/{api → client/api}/index.ts +2 -2
- inspect_ai/_view/www/src/{api → client/api}/types.ts +4 -1
- inspect_ai/_view/www/src/{logfile → client/remote}/remoteLogFile.ts +3 -3
- inspect_ai/_view/www/src/{storage → client/storage}/index.ts +11 -5
- inspect_ai/_view/www/src/components/Card.tsx +1 -1
- inspect_ai/_view/www/src/components/CopyButton.tsx +1 -1
- inspect_ai/_view/www/src/components/DownloadButton.tsx +1 -1
- inspect_ai/_view/www/src/components/ErrorPanel.tsx +1 -1
- inspect_ai/_view/www/src/components/{ExpandablePanel.css → ExpandablePanel.module.css} +14 -11
- inspect_ai/_view/www/src/components/ExpandablePanel.tsx +16 -10
- inspect_ai/_view/www/src/components/FindBand.tsx +1 -1
- inspect_ai/_view/www/src/components/JsonPanel.css +2 -2
- inspect_ai/_view/www/src/components/LargeModal.tsx +12 -1
- inspect_ai/_view/www/src/components/LightboxCarousel.tsx +1 -1
- inspect_ai/_view/www/src/components/MarkdownDiv.tsx +3 -1
- inspect_ai/_view/www/src/components/MessageBand.tsx +1 -1
- inspect_ai/_view/www/src/components/NoContentsPanel.tsx +1 -1
- inspect_ai/_view/www/src/constants.ts +10 -9
- inspect_ai/_view/www/src/index.tsx +27 -11
- inspect_ai/_view/www/src/state/appSlice.ts +44 -5
- inspect_ai/_view/www/src/state/hooks.ts +30 -7
- inspect_ai/_view/www/src/state/logSlice.ts +7 -5
- inspect_ai/_view/www/src/state/logsPolling.ts +1 -1
- inspect_ai/_view/www/src/state/logsSlice.ts +18 -13
- inspect_ai/_view/www/src/state/samplePolling.ts +12 -12
- inspect_ai/_view/www/src/state/sampleSlice.ts +3 -5
- inspect_ai/_view/www/src/state/sampleUtils.ts +1 -1
- inspect_ai/_view/www/src/{scoring/utils.ts → state/scoring.ts} +2 -2
- inspect_ai/_view/www/src/state/store.ts +9 -7
- inspect_ai/_view/www/src/state/utils.ts +1 -1
- inspect_ai/_view/www/src/tests/README.md +49 -0
- inspect_ai/_view/www/src/tests/__mocks__/fileMock.js +1 -0
- inspect_ai/_view/www/src/tests/__mocks__/styleMock.js +1 -0
- inspect_ai/_view/www/src/tests/setupTests.mjs +1 -0
- inspect_ai/_view/www/src/tests/utils/base64.test.ts +23 -0
- inspect_ai/_view/www/src/tests/utils/format.test.ts +127 -0
- inspect_ai/_view/www/src/tests/utils/path.test.ts +54 -0
- inspect_ai/_view/www/src/utils/format.ts +8 -2
- inspect_ai/_view/www/src/utils/path.ts +14 -2
- inspect_ai/_view/www/src/utils/polling.ts +1 -2
- inspect_ai/_view/www/src/utils/uri.ts +32 -0
- inspect_ai/_view/www/yarn.lock +3310 -382
- inspect_ai/agent/_handoff.py +6 -3
- inspect_ai/agent/_human/agent.py +5 -3
- inspect_ai/agent/_human/install.py +16 -7
- inspect_ai/agent/_human/panel.py +14 -1
- inspect_ai/agent/_human/service.py +5 -1
- inspect_ai/agent/_react.py +161 -128
- inspect_ai/agent/_types.py +15 -4
- inspect_ai/approval/_policy.py +2 -2
- inspect_ai/log/_file.py +30 -11
- inspect_ai/log/_log.py +7 -1
- inspect_ai/log/_recorders/eval.py +3 -0
- inspect_ai/log/_recorders/types.py +1 -0
- inspect_ai/log/_samples.py +4 -0
- inspect_ai/model/_call_tools.py +33 -17
- inspect_ai/model/_generate_config.py +10 -2
- inspect_ai/model/_model.py +41 -21
- inspect_ai/model/_model_output.py +2 -1
- inspect_ai/model/_openai.py +10 -8
- inspect_ai/model/_openai_responses.py +95 -42
- inspect_ai/model/_providers/anthropic.py +14 -12
- inspect_ai/model/_providers/google.py +191 -95
- inspect_ai/model/_providers/hf.py +1 -1
- inspect_ai/model/_providers/mistral.py +2 -3
- inspect_ai/model/_providers/openai.py +54 -17
- inspect_ai/model/_providers/openai_o1.py +1 -1
- inspect_ai/model/_providers/openai_responses.py +28 -16
- inspect_ai/model/_providers/openrouter.py +14 -0
- inspect_ai/model/_providers/providers.py +2 -2
- inspect_ai/model/_providers/util/chatapi.py +17 -7
- inspect_ai/model/_providers/vllm.py +1 -1
- inspect_ai/scorer/_metric.py +17 -1
- inspect_ai/scorer/_model.py +51 -6
- inspect_ai/scorer/_scorer.py +1 -1
- inspect_ai/solver/_human_agent.py +3 -0
- inspect_ai/solver/_plan.py +1 -1
- inspect_ai/solver/_solver.py +1 -1
- inspect_ai/solver/_use_tools.py +14 -8
- inspect_ai/tool/__init__.py +16 -1
- inspect_ai/tool/_json_rpc_helpers.py +285 -0
- inspect_ai/tool/_mcp/__init__.py +13 -0
- inspect_ai/tool/_mcp/_context.py +14 -0
- inspect_ai/tool/_mcp/_mcp.py +293 -0
- inspect_ai/tool/_mcp/_sandbox.py +104 -0
- inspect_ai/tool/_mcp/_types.py +31 -0
- inspect_ai/tool/_mcp/connection.py +60 -0
- inspect_ai/tool/_mcp/sampling.py +118 -0
- inspect_ai/tool/_mcp/server.py +112 -0
- inspect_ai/tool/_mcp/tools.py +34 -0
- inspect_ai/tool/_tool.py +13 -0
- inspect_ai/tool/_tool_def.py +24 -7
- inspect_ai/tool/_tool_support_helpers.py +129 -153
- inspect_ai/tool/_tools/_bash_session.py +11 -11
- inspect_ai/tool/_tools/_text_editor.py +6 -6
- inspect_ai/tool/_tools/_web_browser/_web_browser.py +8 -8
- inspect_ai/util/_anyio.py +31 -20
- inspect_ai/util/_json.py +20 -2
- inspect_ai/util/_sandbox/context.py +18 -7
- inspect_ai/util/_sandbox/docker/compose.py +1 -1
- inspect_ai/util/_sandbox/docker/docker.py +92 -21
- inspect_ai/util/_sandbox/environment.py +33 -2
- inspect_ai/util/_sandbox/events.py +2 -2
- inspect_ai/util/_sandbox/service.py +13 -3
- {inspect_ai-0.3.90.dist-info → inspect_ai-0.3.92.dist-info}/METADATA +6 -2
- inspect_ai-0.3.92.dist-info/RECORD +732 -0
- {inspect_ai-0.3.90.dist-info → inspect_ai-0.3.92.dist-info}/WHEEL +1 -1
- inspect_ai/_view/www/src/App.tsx +0 -316
- inspect_ai/_view/www/src/samples/chat/MessageContent.module.css +0 -4
- inspect_ai/_view/www/src/samples/chat/MessageContents.module.css +0 -3
- inspect_ai/_view/www/src/samples/chat/tools/ToolCallView.module.css +0 -3
- inspect_ai/_view/www/src/samples/chat/tools/ToolInput.module.css +0 -14
- inspect_ai/_view/www/src/workspace/WorkSpace.tsx +0 -292
- inspect_ai/_view/www/src/workspace/sidebar/LogDirectoryTitleView.module.css +0 -5
- inspect_ai/_view/www/src/workspace/sidebar/LogDirectoryTitleView.tsx +0 -57
- inspect_ai/_view/www/src/workspace/tabs/JsonTab.tsx +0 -43
- inspect_ai-0.3.90.dist-info/RECORD +0 -705
- /inspect_ai/_view/www/src/{types → @types}/asciicinema-player.d.ts +0 -0
- /inspect_ai/_view/www/src/{types → @types}/jsondiffpatch.d.ts +0 -0
- /inspect_ai/_view/www/src/{types → @types}/markdown-it-katex.d.ts +0 -0
- /inspect_ai/_view/www/src/{types → @types}/prism.d.ts +0 -0
- /inspect_ai/_view/www/src/{appearance → app/appearance}/colors.ts +0 -0
- /inspect_ai/_view/www/src/{appearance → app/appearance}/fonts.ts +0 -0
- /inspect_ai/_view/www/src/{appearance → app/appearance}/styles.ts +0 -0
- /inspect_ai/_view/www/src/{metadata → app/content}/MetaDataGrid.tsx +0 -0
- /inspect_ai/_view/www/src/{metadata → app/content}/MetaDataView.module.css +0 -0
- /inspect_ai/_view/www/src/{metadata → app/content}/MetaDataView.tsx +0 -0
- /inspect_ai/_view/www/src/{metadata → app/content}/MetadataGrid.module.css +0 -0
- /inspect_ai/_view/www/src/{metadata → app/content}/RenderedContent.module.css +0 -0
- /inspect_ai/_view/www/src/{metadata → app/content}/types.ts +0 -0
- /inspect_ai/_view/www/src/{workspace/WorkSpaceView.module.css → app/log-view/LogView.module.css} +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/error/TaskErrorPanel.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/ModelRolesView.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/Navbar.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/PrimaryBar.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/ResultsPanel.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/RunningStatusPanel.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/ScoreGrid.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/SecondaryBar.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/StatusPanel.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/navbar/StatusPanel.tsx +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/tabs/InfoTab.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/tabs/JsonTab.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/tabs/RunningNoSamples.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/tabs/RunningNoSamples.tsx +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/types.ts +0 -0
- /inspect_ai/_view/www/src/{workspace → app/log-view}/utils.ts +0 -0
- /inspect_ai/_view/www/src/{plan → app/plan}/DatasetDetailView.module.css +0 -0
- /inspect_ai/_view/www/src/{plan → app/plan}/DetailStep.module.css +0 -0
- /inspect_ai/_view/www/src/{plan → app/plan}/ModelCard.module.css +0 -0
- /inspect_ai/_view/www/src/{plan → app/plan}/PlanDetailView.module.css +0 -0
- /inspect_ai/_view/www/src/{plan → app/plan}/ScorerDetailView.module.css +0 -0
- /inspect_ai/_view/www/src/{plan → app/plan}/ScorerDetailView.tsx +0 -0
- /inspect_ai/_view/www/src/{plan → app/plan}/SolverDetailView.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/InlineSampleDisplay.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/chat/ChatMessageRow.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/chat/ChatViewVirtualList.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/descriptor/score/BooleanScoreDescriptor.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/descriptor/score/ObjectScoreDescriptor.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/descriptor/score/PassFailScoreDescriptor.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/error/FlatSampleErrorView.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/error/FlatSampleErrorView.tsx +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/error/SampleErrorView.tsx +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/error/error.ts +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/list/SampleFooter.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/list/SampleFooter.tsx +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/list/SampleHeader.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/list/SampleList.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/list/SampleSeparator.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/list/SampleSeparator.tsx +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/sample-tools/EpochFilter.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/sample-tools/EpochFilter.tsx +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/sample-tools/SelectScorer.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/sample-tools/SortFilter.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/sample-tools/sample-filter/SampleFilter.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/sample-tools/sample-filter/tokenize.ts +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/scores/SampleScores.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/scores/SampleScoresGrid.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/scores/SampleScoresView.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/InfoEventView.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/LoggerEventView.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/SampleInitEventView.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/SandboxEventView.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/ScoreEventView.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/SubtaskEventView.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/ToolEventView.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventNav.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventNav.tsx +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventNavs.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventNavs.tsx +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventPanel.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventProgressPanel.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventProgressPanel.tsx +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventRow.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventRow.tsx +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventSection.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventSection.tsx +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/event/EventTimingPanel.module.css +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/state/StateDiffView.tsx +0 -0
- /inspect_ai/_view/www/src/{samples → app/samples}/transcript/state/StateEventView.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app}/sidebar/EvalStatus.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app}/sidebar/SidebarLogEntry.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app}/sidebar/SidebarScoreView.module.css +0 -0
- /inspect_ai/_view/www/src/{workspace → app}/sidebar/SidebarScoresView.module.css +0 -0
- /inspect_ai/_view/www/src/{usage → app/usage}/ModelUsagePanel.module.css +0 -0
- /inspect_ai/_view/www/src/{usage → app/usage}/TokenTable.module.css +0 -0
- /inspect_ai/_view/www/src/{usage → app/usage}/UsageCard.module.css +0 -0
- /inspect_ai/_view/www/src/{api → client/api}/api-shared.ts +0 -0
- /inspect_ai/_view/www/src/{api → client/api}/jsonrpc.ts +0 -0
- /inspect_ai/_view/www/src/{logfile → client/remote}/remoteZipFile.ts +0 -0
- {inspect_ai-0.3.90.dist-info → inspect_ai-0.3.92.dist-info}/entry_points.txt +0 -0
- {inspect_ai-0.3.90.dist-info → inspect_ai-0.3.92.dist-info}/licenses/LICENSE +0 -0
- {inspect_ai-0.3.90.dist-info → inspect_ai-0.3.92.dist-info}/top_level.txt +0 -0
@@ -1,12 +1,12 @@
|
|
1
|
-
import { TabPanel, TabSet } from "
|
2
|
-
import { MetaDataView } from "../
|
1
|
+
import { TabPanel, TabSet } from "../../components/TabSet";
|
2
|
+
import { MetaDataView } from "../content/MetaDataView";
|
3
3
|
|
4
|
-
import { escapeSelector } from "
|
5
|
-
import { isVscode } from "
|
4
|
+
import { escapeSelector } from "../../utils/html";
|
5
|
+
import { isVscode } from "../../utils/vscode";
|
6
6
|
|
7
|
+
import { ANSIDisplay } from "../../components/AnsiDisplay";
|
8
|
+
import { ToolButton } from "../../components/ToolButton";
|
7
9
|
import { ApplicationIcons } from "../appearance/icons";
|
8
|
-
import { ANSIDisplay } from "../components/AnsiDisplay";
|
9
|
-
import { ToolButton } from "../components/ToolButton";
|
10
10
|
|
11
11
|
import clsx from "clsx";
|
12
12
|
import {
|
@@ -17,10 +17,12 @@ import {
|
|
17
17
|
useCallback,
|
18
18
|
useMemo,
|
19
19
|
} from "react";
|
20
|
-
import {
|
21
|
-
import {
|
22
|
-
import {
|
23
|
-
import {
|
20
|
+
import { useNavigate, useParams } from "react-router-dom";
|
21
|
+
import { EvalSample, Events } from "../../@types/log";
|
22
|
+
import { SampleSummary } from "../../client/api/types";
|
23
|
+
import { Card, CardBody, CardHeader } from "../../components/Card";
|
24
|
+
import { JSONPanel } from "../../components/JsonPanel";
|
25
|
+
import { NoContentsPanel } from "../../components/NoContentsPanel";
|
24
26
|
import {
|
25
27
|
kSampleErrorTabId,
|
26
28
|
kSampleJsonTabId,
|
@@ -28,13 +30,13 @@ import {
|
|
28
30
|
kSampleMetdataTabId,
|
29
31
|
kSampleScoringTabId,
|
30
32
|
kSampleTranscriptTabId,
|
31
|
-
} from "
|
32
|
-
import {
|
33
|
-
import { useStore } from "
|
34
|
-
import {
|
33
|
+
} from "../../constants";
|
34
|
+
import { useFilteredSamples, useSampleData } from "../../state/hooks";
|
35
|
+
import { useStore } from "../../state/store";
|
36
|
+
import { formatTime } from "../../utils/format";
|
37
|
+
import { printHeadingHtml, printHtml } from "../../utils/print";
|
38
|
+
import { sampleUrl } from "../routing/url";
|
35
39
|
import { ModelTokenTable } from "../usage/ModelTokenTable";
|
36
|
-
import { formatTime } from "../utils/format";
|
37
|
-
import { printHeadingHtml, printHtml } from "../utils/print";
|
38
40
|
import { ChatViewVirtualList } from "./chat/ChatViewVirtualList";
|
39
41
|
import { messagesFromEvents } from "./chat/messages";
|
40
42
|
import styles from "./SampleDisplay.module.css";
|
@@ -44,32 +46,38 @@ import { TranscriptVirtualList } from "./transcript/TranscriptView";
|
|
44
46
|
|
45
47
|
interface SampleDisplayProps {
|
46
48
|
id: string;
|
47
|
-
sample?: EvalSample;
|
48
|
-
selectedTab?: string;
|
49
|
-
setSelectedTab: (tab: string) => void;
|
50
49
|
scrollRef: RefObject<HTMLDivElement | null>;
|
51
|
-
runningEvents?: Events;
|
52
50
|
}
|
53
51
|
|
54
52
|
/**
|
55
53
|
* Component to display a sample with relevant context and visibility control.
|
56
54
|
*/
|
57
|
-
export const SampleDisplay: FC<SampleDisplayProps> = ({
|
58
|
-
id,
|
59
|
-
sample,
|
60
|
-
selectedTab,
|
61
|
-
setSelectedTab,
|
62
|
-
scrollRef,
|
63
|
-
runningEvents: runningSampleData,
|
64
|
-
}) => {
|
55
|
+
export const SampleDisplay: FC<SampleDisplayProps> = ({ id, scrollRef }) => {
|
65
56
|
// Tab ids
|
66
57
|
const baseId = `sample-dialog`;
|
67
|
-
const
|
58
|
+
const filteredSamples = useFilteredSamples();
|
68
59
|
const selectedSampleIndex = useStore(
|
69
60
|
(state) => state.log.selectedSampleIndex,
|
70
61
|
);
|
71
62
|
|
72
|
-
const
|
63
|
+
const sampleData = useSampleData();
|
64
|
+
const sample = sampleData.sample;
|
65
|
+
const runningSampleData = sampleData.running;
|
66
|
+
|
67
|
+
// Selected tab handling
|
68
|
+
const selectedTab = useStore((state) => state.app.tabs.sample);
|
69
|
+
const setSelectedTab = useStore((state) => state.appActions.setSampleTab);
|
70
|
+
|
71
|
+
// Get sample tab from URL if available
|
72
|
+
const { sampleTabId } = useParams<{ sampleTabId?: string }>();
|
73
|
+
|
74
|
+
// Use sampleTabId from URL if available, otherwise use the one from state
|
75
|
+
const effectiveSelectedTab = sampleTabId || selectedTab;
|
76
|
+
|
77
|
+
// Navigation hook for URL updates
|
78
|
+
const navigate = useNavigate();
|
79
|
+
|
80
|
+
const sampleSummary = filteredSamples[selectedSampleIndex];
|
73
81
|
|
74
82
|
// Consolidate the events and messages into the proper list
|
75
83
|
// whether running or not
|
@@ -84,13 +92,42 @@ export const SampleDisplay: FC<SampleDisplayProps> = ({
|
|
84
92
|
}
|
85
93
|
}, [sample?.messages, runningSampleData]);
|
86
94
|
|
95
|
+
// Get all URL parameters at component level
|
96
|
+
const {
|
97
|
+
logPath: urlLogPath,
|
98
|
+
tabId: urlTabId,
|
99
|
+
sampleId: urlSampleId,
|
100
|
+
epoch: urlEpoch,
|
101
|
+
} = useParams<{
|
102
|
+
logPath?: string;
|
103
|
+
tabId?: string;
|
104
|
+
sampleId?: string;
|
105
|
+
epoch?: string;
|
106
|
+
}>();
|
107
|
+
|
87
108
|
// Tab selection
|
88
|
-
const onSelectedTab = (
|
89
|
-
|
90
|
-
|
91
|
-
|
92
|
-
|
93
|
-
|
109
|
+
const onSelectedTab = useCallback(
|
110
|
+
(e: MouseEvent<HTMLElement>) => {
|
111
|
+
const el = e.currentTarget as HTMLElement;
|
112
|
+
const id = el.id;
|
113
|
+
setSelectedTab(id);
|
114
|
+
|
115
|
+
// Use navigation hook to update URL with tab
|
116
|
+
if (id !== sampleTabId && urlLogPath) {
|
117
|
+
const url = sampleUrl(urlLogPath, urlSampleId, urlEpoch, id);
|
118
|
+
navigate(url);
|
119
|
+
}
|
120
|
+
},
|
121
|
+
[
|
122
|
+
sampleTabId,
|
123
|
+
urlLogPath,
|
124
|
+
urlTabId,
|
125
|
+
urlSampleId,
|
126
|
+
urlEpoch,
|
127
|
+
navigate,
|
128
|
+
setSelectedTab,
|
129
|
+
],
|
130
|
+
);
|
94
131
|
|
95
132
|
const sampleMetadatas = metadataViewsForSample(`${baseId}-${id}`, sample);
|
96
133
|
|
@@ -134,7 +171,8 @@ export const SampleDisplay: FC<SampleDisplayProps> = ({
|
|
134
171
|
title="Transcript"
|
135
172
|
onSelected={onSelectedTab}
|
136
173
|
selected={
|
137
|
-
|
174
|
+
effectiveSelectedTab === kSampleTranscriptTabId ||
|
175
|
+
effectiveSelectedTab === undefined
|
138
176
|
}
|
139
177
|
scrollable={false}
|
140
178
|
>
|
@@ -152,7 +190,7 @@ export const SampleDisplay: FC<SampleDisplayProps> = ({
|
|
152
190
|
className={clsx("sample-tab", styles.fullWidth, styles.chat)}
|
153
191
|
title="Messages"
|
154
192
|
onSelected={onSelectedTab}
|
155
|
-
selected={
|
193
|
+
selected={effectiveSelectedTab === kSampleMessagesTabId}
|
156
194
|
scrollable={false}
|
157
195
|
>
|
158
196
|
<ChatViewVirtualList
|
@@ -171,7 +209,7 @@ export const SampleDisplay: FC<SampleDisplayProps> = ({
|
|
171
209
|
className="sample-tab"
|
172
210
|
title="Scoring"
|
173
211
|
onSelected={onSelectedTab}
|
174
|
-
selected={
|
212
|
+
selected={effectiveSelectedTab === kSampleScoringTabId}
|
175
213
|
>
|
176
214
|
<SampleScoresView sample={sample} />
|
177
215
|
</TabPanel>
|
@@ -180,7 +218,7 @@ export const SampleDisplay: FC<SampleDisplayProps> = ({
|
|
180
218
|
className={clsx("sample-tab")}
|
181
219
|
title="Metadata"
|
182
220
|
onSelected={onSelectedTab}
|
183
|
-
selected={
|
221
|
+
selected={effectiveSelectedTab === kSampleMetdataTabId}
|
184
222
|
>
|
185
223
|
{sampleMetadatas.length > 0 ? (
|
186
224
|
<div className={clsx(styles.metadataPanel)}>{sampleMetadatas}</div>
|
@@ -188,19 +226,48 @@ export const SampleDisplay: FC<SampleDisplayProps> = ({
|
|
188
226
|
<NoContentsPanel text="No metadata" />
|
189
227
|
)}
|
190
228
|
</TabPanel>
|
191
|
-
{sample?.error
|
229
|
+
{sample?.error ||
|
230
|
+
(sample?.error_retries && sample?.error_retries.length > 0) ? (
|
192
231
|
<TabPanel
|
193
232
|
id={kSampleErrorTabId}
|
194
233
|
className="sample-tab"
|
195
|
-
title="
|
234
|
+
title="Errors"
|
196
235
|
onSelected={onSelectedTab}
|
197
|
-
selected={
|
236
|
+
selected={effectiveSelectedTab === kSampleErrorTabId}
|
198
237
|
>
|
199
|
-
<div className={clsx(styles.
|
200
|
-
|
201
|
-
|
202
|
-
|
203
|
-
|
238
|
+
<div className={clsx(styles.error)}>
|
239
|
+
{sample?.error ? (
|
240
|
+
<Card key={`sample-error}`}>
|
241
|
+
<CardHeader label={`Sample Error`} />
|
242
|
+
<CardBody>
|
243
|
+
<ANSIDisplay
|
244
|
+
output={sample.error.traceback_ansi}
|
245
|
+
className={clsx("text-size-small", styles.ansi)}
|
246
|
+
style={{
|
247
|
+
fontSize: "clamp(0.4rem, calc(0.15em + 1vw), 0.8rem)",
|
248
|
+
margin: "0.5em 0",
|
249
|
+
}}
|
250
|
+
/>
|
251
|
+
</CardBody>
|
252
|
+
</Card>
|
253
|
+
) : undefined}
|
254
|
+
{sample.error_retries?.map((retry, index) => {
|
255
|
+
return (
|
256
|
+
<Card key={`sample-retry-error-${index}`}>
|
257
|
+
<CardHeader label={`Attempt ${index + 1}`} />
|
258
|
+
<CardBody>
|
259
|
+
<ANSIDisplay
|
260
|
+
output={retry.traceback_ansi}
|
261
|
+
className={clsx("text-size-small", styles.ansi)}
|
262
|
+
style={{
|
263
|
+
fontSize: "clamp(0.4rem, calc(0.15em + 1vw), 0.8rem)",
|
264
|
+
margin: "0.5em 0",
|
265
|
+
}}
|
266
|
+
/>
|
267
|
+
</CardBody>
|
268
|
+
</Card>
|
269
|
+
);
|
270
|
+
})}
|
204
271
|
</div>
|
205
272
|
</TabPanel>
|
206
273
|
) : null}
|
@@ -209,7 +276,7 @@ export const SampleDisplay: FC<SampleDisplayProps> = ({
|
|
209
276
|
className={"sample-tab"}
|
210
277
|
title="JSON"
|
211
278
|
onSelected={onSelectedTab}
|
212
|
-
selected={
|
279
|
+
selected={effectiveSelectedTab === kSampleJsonTabId}
|
213
280
|
>
|
214
281
|
{!sample ? (
|
215
282
|
<NoContentsPanel text="JSON not available" />
|
@@ -1,12 +1,12 @@
|
|
1
1
|
import clsx from "clsx";
|
2
|
-
import {
|
3
|
-
import {
|
4
|
-
import { arrayToString, formatTime, inputString } from "
|
2
|
+
import { EvalSample, Target, TotalTime, WorkingTime } from "../../@types/log";
|
3
|
+
import { MarkdownDiv } from "../../components/MarkdownDiv";
|
4
|
+
import { arrayToString, formatTime, inputString } from "../../utils/format";
|
5
5
|
import { FlatSampleError } from "./error/FlatSampleErrorView";
|
6
6
|
|
7
7
|
import { FC, ReactNode } from "react";
|
8
|
-
import { SampleSummary } from "
|
9
|
-
import { useSampleDescriptor, useScore } from "
|
8
|
+
import { SampleSummary } from "../../client/api/types";
|
9
|
+
import { useSampleDescriptor, useScore } from "../../state/hooks";
|
10
10
|
import styles from "./SampleSummaryView.module.css";
|
11
11
|
import { SamplesDescriptor } from "./descriptor/samplesDescriptor";
|
12
12
|
|
@@ -30,6 +30,7 @@ interface SampleFields {
|
|
30
30
|
target: Target;
|
31
31
|
answer?: string;
|
32
32
|
limit?: string;
|
33
|
+
retries?: number;
|
33
34
|
working_time?: WorkingTime;
|
34
35
|
total_time?: TotalTime;
|
35
36
|
error?: string;
|
@@ -38,7 +39,7 @@ interface SampleFields {
|
|
38
39
|
function isEvalSample(
|
39
40
|
sample: SampleSummary | EvalSample,
|
40
41
|
): sample is EvalSample {
|
41
|
-
return "
|
42
|
+
return "store" in sample;
|
42
43
|
}
|
43
44
|
|
44
45
|
const resolveSample = (
|
@@ -64,6 +65,9 @@ const resolveSample = (
|
|
64
65
|
const working_time = isEvalSample(sample) ? sample.working_time : undefined;
|
65
66
|
const total_time = isEvalSample(sample) ? sample.total_time : undefined;
|
66
67
|
const error = isEvalSample(sample) ? sample.error?.message : undefined;
|
68
|
+
const retries = isEvalSample(sample)
|
69
|
+
? sample.error_retries?.length
|
70
|
+
: sample.retries;
|
67
71
|
|
68
72
|
return {
|
69
73
|
id: sample.id,
|
@@ -71,6 +75,7 @@ const resolveSample = (
|
|
71
75
|
target,
|
72
76
|
answer,
|
73
77
|
limit,
|
78
|
+
retries,
|
74
79
|
working_time,
|
75
80
|
total_time,
|
76
81
|
error,
|
@@ -91,23 +96,12 @@ export const SampleSummaryView: FC<SampleSummaryViewProps> = ({
|
|
91
96
|
}
|
92
97
|
const fields = resolveSample(sample, sampleDescriptor);
|
93
98
|
|
94
|
-
const input =
|
95
|
-
sampleDescriptor?.messageShape.normalized.input > 0
|
96
|
-
? Math.max(0.15, sampleDescriptor.messageShape.normalized.input)
|
97
|
-
: 0;
|
98
|
-
const target =
|
99
|
-
sampleDescriptor?.messageShape.normalized.target > 0
|
100
|
-
? Math.max(0.15, sampleDescriptor.messageShape.normalized.target)
|
101
|
-
: 0;
|
102
|
-
const answer =
|
103
|
-
sampleDescriptor?.messageShape.normalized.answer > 0
|
104
|
-
? Math.max(0.15, sampleDescriptor.messageShape.normalized.answer)
|
105
|
-
: 0;
|
106
99
|
const limitSize =
|
107
100
|
sampleDescriptor?.messageShape.normalized.limit > 0
|
108
101
|
? Math.max(0.15, sampleDescriptor.messageShape.normalized.limit)
|
109
102
|
: 0;
|
110
|
-
const
|
103
|
+
const retrySize =
|
104
|
+
sampleDescriptor?.messageShape.normalized.retries > 0 ? 6 : 0;
|
111
105
|
const idSize = Math.max(
|
112
106
|
2,
|
113
107
|
Math.min(10, sampleDescriptor?.messageShape.raw.id),
|
@@ -124,7 +118,7 @@ export const SampleSummaryView: FC<SampleSummaryViewProps> = ({
|
|
124
118
|
columns.push({
|
125
119
|
label: "Input",
|
126
120
|
value: <MarkdownDiv markdown={fields.input.join(" ")} />,
|
127
|
-
size:
|
121
|
+
size: `minmax(auto, 5fr)`,
|
128
122
|
clamp: true,
|
129
123
|
});
|
130
124
|
|
@@ -137,7 +131,7 @@ export const SampleSummaryView: FC<SampleSummaryViewProps> = ({
|
|
137
131
|
className={clsx("no-last-para-padding", styles.target)}
|
138
132
|
/>
|
139
133
|
),
|
140
|
-
size:
|
134
|
+
size: `minmax(auto, 3fr)`,
|
141
135
|
clamp: true,
|
142
136
|
});
|
143
137
|
}
|
@@ -147,13 +141,13 @@ export const SampleSummaryView: FC<SampleSummaryViewProps> = ({
|
|
147
141
|
label: "Answer",
|
148
142
|
value: sample ? (
|
149
143
|
<MarkdownDiv
|
150
|
-
markdown={fields.answer}
|
144
|
+
markdown={fields.answer || ""}
|
151
145
|
className={clsx("no-last-para-padding", styles.answer)}
|
152
146
|
/>
|
153
147
|
) : (
|
154
148
|
""
|
155
149
|
),
|
156
|
-
size:
|
150
|
+
size: `minmax(auto, 5fr)`,
|
157
151
|
clamp: true,
|
158
152
|
});
|
159
153
|
}
|
@@ -169,7 +163,7 @@ export const SampleSummaryView: FC<SampleSummaryViewProps> = ({
|
|
169
163
|
columns.push({
|
170
164
|
label: "Time",
|
171
165
|
value: formatTime(fields.total_time),
|
172
|
-
size:
|
166
|
+
size: `fit-content(10rem)`,
|
173
167
|
center: true,
|
174
168
|
title: toolTip(fields.working_time),
|
175
169
|
});
|
@@ -179,7 +173,16 @@ export const SampleSummaryView: FC<SampleSummaryViewProps> = ({
|
|
179
173
|
columns.push({
|
180
174
|
label: "Limit",
|
181
175
|
value: fields.limit,
|
182
|
-
size:
|
176
|
+
size: `fit-content(10rem)`,
|
177
|
+
center: true,
|
178
|
+
});
|
179
|
+
}
|
180
|
+
|
181
|
+
if (fields?.retries && retrySize > 0) {
|
182
|
+
columns.push({
|
183
|
+
label: "Retries",
|
184
|
+
value: fields.retries,
|
185
|
+
size: `fit-content(${retrySize}rem)`,
|
183
186
|
center: true,
|
184
187
|
});
|
185
188
|
}
|
@@ -192,7 +195,7 @@ export const SampleSummaryView: FC<SampleSummaryViewProps> = ({
|
|
192
195
|
sampleDescriptor?.evalDescriptor.score(sample, currentScore)?.render() ||
|
193
196
|
""
|
194
197
|
),
|
195
|
-
size: "
|
198
|
+
size: "fit-content(15em)",
|
196
199
|
center: true,
|
197
200
|
});
|
198
201
|
|
@@ -1,8 +1,8 @@
|
|
1
1
|
import { FC } from "react";
|
2
2
|
import { Fragment } from "react/jsx-runtime";
|
3
|
-
import { SampleSummary } from "
|
4
|
-
import { useScore, useScores } from "
|
5
|
-
import { useStore } from "
|
3
|
+
import { SampleSummary } from "../../client/api/types";
|
4
|
+
import { useScore, useScores } from "../../state/hooks";
|
5
|
+
import { useStore } from "../../state/store";
|
6
6
|
import { EpochFilter } from "./sample-tools/EpochFilter";
|
7
7
|
import { SampleFilter } from "./sample-tools/sample-filter/SampleFilter";
|
8
8
|
import { SelectScorer } from "./sample-tools/SelectScorer";
|
@@ -1,11 +1,14 @@
|
|
1
1
|
.message {
|
2
2
|
font-weight: 300;
|
3
|
-
padding-bottom: 0.5em;
|
4
3
|
margin-left: 0;
|
5
4
|
margin-right: 0;
|
6
5
|
white-space: normal;
|
7
6
|
}
|
8
7
|
|
8
|
+
.padded {
|
9
|
+
padding-bottom: 1em;
|
10
|
+
}
|
11
|
+
|
9
12
|
.systemRole {
|
10
13
|
opacity: 0.7;
|
11
14
|
}
|
@@ -15,7 +18,7 @@
|
|
15
18
|
grid-template-columns: max-content auto;
|
16
19
|
column-gap: 0.3em;
|
17
20
|
font-weight: 500;
|
18
|
-
margin-bottom: 0.
|
21
|
+
margin-bottom: 0.3em;
|
19
22
|
}
|
20
23
|
|
21
24
|
.messageContents {
|
@@ -1,22 +1,24 @@
|
|
1
1
|
import clsx from "clsx";
|
2
2
|
import { FC } from "react";
|
3
|
-
import ExpandablePanel from "../../components/ExpandablePanel";
|
4
3
|
import {
|
5
4
|
ChatMessageAssistant,
|
6
5
|
ChatMessageSystem,
|
7
6
|
ChatMessageTool,
|
8
7
|
ChatMessageUser,
|
9
|
-
} from "
|
8
|
+
} from "../../../@types/log";
|
9
|
+
import ExpandablePanel from "../../../components/ExpandablePanel";
|
10
10
|
import styles from "./ChatMessage.module.css";
|
11
11
|
import { MessageContents } from "./MessageContents";
|
12
12
|
import { iconForMsg } from "./messages";
|
13
|
+
import { ChatViewToolCallStyle } from "./types";
|
13
14
|
|
14
15
|
interface ChatMessageProps {
|
15
16
|
id: string;
|
16
17
|
message: ChatMessageAssistant | ChatMessageSystem | ChatMessageUser;
|
17
18
|
toolMessages: ChatMessageTool[];
|
18
19
|
indented?: boolean;
|
19
|
-
toolCallStyle:
|
20
|
+
toolCallStyle: ChatViewToolCallStyle;
|
21
|
+
padded?: boolean;
|
20
22
|
}
|
21
23
|
|
22
24
|
export const ChatMessage: FC<ChatMessageProps> = ({
|
@@ -25,6 +27,7 @@ export const ChatMessage: FC<ChatMessageProps> = ({
|
|
25
27
|
toolMessages,
|
26
28
|
indented,
|
27
29
|
toolCallStyle,
|
30
|
+
padded,
|
28
31
|
}) => {
|
29
32
|
const collapse = message.role === "system" || message.role === "user";
|
30
33
|
return (
|
@@ -33,6 +36,7 @@ export const ChatMessage: FC<ChatMessageProps> = ({
|
|
33
36
|
message.role,
|
34
37
|
"text-size-base",
|
35
38
|
styles.message,
|
39
|
+
padded ? styles.padded : undefined,
|
36
40
|
message.role === "system" ? styles.systemRole : undefined,
|
37
41
|
)}
|
38
42
|
>
|
@@ -46,7 +50,11 @@ export const ChatMessage: FC<ChatMessageProps> = ({
|
|
46
50
|
indented ? styles.indented : undefined,
|
47
51
|
)}
|
48
52
|
>
|
49
|
-
<ExpandablePanel
|
53
|
+
<ExpandablePanel
|
54
|
+
id={`${id}-message`}
|
55
|
+
collapse={collapse}
|
56
|
+
lines={collapse ? 15 : 25}
|
57
|
+
>
|
50
58
|
<MessageContents
|
51
59
|
id={`${id}-contents`}
|
52
60
|
key={`${id}-contents`}
|
@@ -1,12 +1,12 @@
|
|
1
1
|
import { FC } from "react";
|
2
|
-
import { NavPills } from "../../components/NavPills.tsx";
|
3
|
-
import { Buckets, ContentRenderer } from "../../metadata/types.ts";
|
4
2
|
import {
|
5
3
|
ChatMessageAssistant,
|
6
4
|
ChatMessageSystem,
|
7
5
|
ChatMessageTool,
|
8
6
|
ChatMessageUser,
|
9
|
-
} from "
|
7
|
+
} from "../../../@types/log";
|
8
|
+
import { NavPills } from "../../../components/NavPills.tsx";
|
9
|
+
import { Buckets, ContentRenderer } from "../../content/types.ts";
|
10
10
|
import { ChatView } from "./ChatView";
|
11
11
|
|
12
12
|
/**
|
@@ -4,13 +4,15 @@ import { ChatMessage } from "./ChatMessage";
|
|
4
4
|
import { FC } from "react";
|
5
5
|
import styles from "./ChatMessageRow.module.css";
|
6
6
|
import { ResolvedMessage } from "./messages";
|
7
|
+
import { ChatViewToolCallStyle } from "./types";
|
7
8
|
|
8
9
|
interface ChatMessageRowProps {
|
9
10
|
parentName: string;
|
10
11
|
number?: number;
|
11
12
|
resolvedMessage: ResolvedMessage;
|
12
|
-
toolCallStyle:
|
13
|
+
toolCallStyle: ChatViewToolCallStyle;
|
13
14
|
indented?: boolean;
|
15
|
+
padded?: boolean;
|
14
16
|
}
|
15
17
|
|
16
18
|
/**
|
@@ -22,6 +24,7 @@ export const ChatMessageRow: FC<ChatMessageRowProps> = ({
|
|
22
24
|
resolvedMessage,
|
23
25
|
toolCallStyle,
|
24
26
|
indented,
|
27
|
+
padded,
|
25
28
|
}) => {
|
26
29
|
if (number) {
|
27
30
|
return (
|
@@ -41,6 +44,7 @@ export const ChatMessageRow: FC<ChatMessageRowProps> = ({
|
|
41
44
|
toolMessages={resolvedMessage.toolMessages}
|
42
45
|
indented={indented}
|
43
46
|
toolCallStyle={toolCallStyle}
|
47
|
+
padded={padded}
|
44
48
|
/>
|
45
49
|
</div>
|
46
50
|
);
|
@@ -52,6 +56,7 @@ export const ChatMessageRow: FC<ChatMessageRowProps> = ({
|
|
52
56
|
toolMessages={resolvedMessage.toolMessages}
|
53
57
|
indented={indented}
|
54
58
|
toolCallStyle={toolCallStyle}
|
59
|
+
padded={padded}
|
55
60
|
/>
|
56
61
|
);
|
57
62
|
}
|
@@ -1,13 +1,14 @@
|
|
1
1
|
import clsx from "clsx";
|
2
2
|
import { FC } from "react";
|
3
|
-
import { Messages } from "
|
3
|
+
import { Messages } from "../../../@types/log";
|
4
4
|
import { ChatMessageRow } from "./ChatMessageRow";
|
5
5
|
import { resolveMessages } from "./messages";
|
6
|
+
import { ChatViewToolCallStyle } from "./types";
|
6
7
|
|
7
8
|
interface ChatViewProps {
|
8
9
|
id?: string;
|
9
10
|
messages: Messages;
|
10
|
-
toolCallStyle?:
|
11
|
+
toolCallStyle?: ChatViewToolCallStyle;
|
11
12
|
title?: string;
|
12
13
|
indented?: boolean;
|
13
14
|
numbered?: boolean;
|
@@ -39,6 +40,7 @@ export const ChatView: FC<ChatViewProps> = ({
|
|
39
40
|
resolvedMessage={msg}
|
40
41
|
indented={indented}
|
41
42
|
toolCallStyle={toolCallStyle}
|
43
|
+
padded={index < collapsedMessages.length - 1}
|
42
44
|
/>
|
43
45
|
);
|
44
46
|
})}
|
@@ -1,16 +1,17 @@
|
|
1
1
|
import { FC, memo, ReactNode, RefObject, useMemo } from "react";
|
2
|
-
import { Messages } from "
|
2
|
+
import { Messages } from "../../../@types/log";
|
3
3
|
|
4
4
|
import { ChatMessageRow } from "./ChatMessageRow";
|
5
5
|
import { ResolvedMessage, resolveMessages } from "./messages";
|
6
6
|
|
7
|
-
import { LiveVirtualList } from "
|
7
|
+
import { LiveVirtualList } from "../../../components/LiveVirtualList";
|
8
|
+
import { ChatViewToolCallStyle } from "./types";
|
8
9
|
|
9
10
|
interface ChatViewVirtualListProps {
|
10
11
|
id: string;
|
11
12
|
className?: string | string[];
|
12
13
|
messages: Messages;
|
13
|
-
toolCallStyle:
|
14
|
+
toolCallStyle: ChatViewToolCallStyle;
|
14
15
|
indented: boolean;
|
15
16
|
numbered?: boolean;
|
16
17
|
scrollRef?: RefObject<HTMLDivElement | null>;
|
@@ -46,6 +47,7 @@ export const ChatViewVirtualList: FC<ChatViewVirtualListProps> = memo(
|
|
46
47
|
resolvedMessage={item}
|
47
48
|
indented={indented}
|
48
49
|
toolCallStyle={toolCallStyle}
|
50
|
+
padded={index < collapsedMessages.length - 1}
|
49
51
|
/>
|
50
52
|
);
|
51
53
|
};
|
@@ -0,0 +1,12 @@
|
|
1
|
+
.contentImage {
|
2
|
+
max-width: 800px;
|
3
|
+
border: solid var(--bs-border-color) 1px;
|
4
|
+
}
|
5
|
+
|
6
|
+
.reasoning {
|
7
|
+
border: solid var(--bs-light-border-subtle) 1px;
|
8
|
+
padding: 1em;
|
9
|
+
margin-bottom: 0.5em;
|
10
|
+
background-color: var(--bs-light-bg-subtle);
|
11
|
+
border-radius: var(--bs-border-radius);
|
12
|
+
}
|