@tangle-network/agent-eval 0.220.2 → 0.221.1
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.
- package/CHANGELOG.md +64 -0
- package/dist/adapters/http.d.ts +2 -2
- package/dist/{agent-profile-D01Io2Xw.d.ts → agent-profile-Bc7ADJY4.d.ts} +2 -2
- package/dist/{agent-profile-D01Io2Xw.d.ts.map → agent-profile-Bc7ADJY4.d.ts.map} +1 -1
- package/dist/analyst/index.d.ts +77 -78
- package/dist/analyst/index.d.ts.map +1 -1
- package/dist/analyst/index.js +204 -25
- package/dist/analyst/index.js.map +1 -1
- package/dist/{analyze-runs-DqPJaJ4X.js → analyze-runs-CHmsykNJ.js} +4 -4
- package/dist/{analyze-runs-DqPJaJ4X.js.map → analyze-runs-CHmsykNJ.js.map} +1 -1
- package/dist/{analyze-runs-BWAHBDAw.d.ts → analyze-runs-D7In4c3z.d.ts} +2 -2
- package/dist/{analyze-runs-BWAHBDAw.d.ts.map → analyze-runs-D7In4c3z.d.ts.map} +1 -1
- package/dist/{benchmark-command-Dx0E4a6z.js → benchmark-command-DiGqnGUt.js} +13 -11
- package/dist/benchmark-command-DiGqnGUt.js.map +1 -0
- package/dist/benchmarks/index.d.ts +4 -4
- package/dist/benchmarks/index.js +3 -3
- package/dist/calibration-CTwxpJ-t.js +197 -0
- package/dist/calibration-CTwxpJ-t.js.map +1 -0
- package/dist/campaign/index.d.ts +6 -6
- package/dist/campaign/index.js +5 -5
- package/dist/{campaign-BVPC6ODQ.js → campaign-Bt6rbWIU.js} +9 -8
- package/dist/{campaign-BVPC6ODQ.js.map → campaign-Bt6rbWIU.js.map} +1 -1
- package/dist/cli.js +1 -1
- package/dist/{completion-verifier-Cigap8CG.js → completion-verifier-YwoeujPl.js} +2 -2
- package/dist/completion-verifier-YwoeujPl.js.map +1 -0
- package/dist/contract/index.d.ts +10 -10
- package/dist/contract/index.js +7 -7
- package/dist/decision-point-Co3u3W8k.js +831 -0
- package/dist/decision-point-Co3u3W8k.js.map +1 -0
- package/dist/default-registry-BMHqNR6_.d.ts +257 -0
- package/dist/default-registry-BMHqNR6_.d.ts.map +1 -0
- package/dist/{default-registry-ovrIBGmH.js → default-registry-BbFCkqvB.js} +607 -14
- package/dist/default-registry-BbFCkqvB.js.map +1 -0
- package/dist/{define-agent-eval-C2nFhHPu.d.ts → define-agent-eval-Bsoi2GUV.d.ts} +7 -7
- package/dist/{define-agent-eval-C2nFhHPu.d.ts.map → define-agent-eval-Bsoi2GUV.d.ts.map} +1 -1
- package/dist/{define-agent-eval-UQY3VbJ2.js → define-agent-eval-CJhECJNL.js} +4 -4
- package/dist/{define-agent-eval-UQY3VbJ2.js.map → define-agent-eval-CJhECJNL.js.map} +1 -1
- package/dist/diagnosis/index.d.ts +2 -2
- package/dist/diagnosis/index.js +1 -1
- package/dist/{dspy-rlm-engine-Bn_Mt_LW.js → dspy-rlm-engine-BTy7u0jQ.js} +3 -3
- package/dist/{dspy-rlm-engine-Bn_Mt_LW.js.map → dspy-rlm-engine-BTy7u0jQ.js.map} +1 -1
- package/dist/{engine-BpGGw6cd.d.ts → engine-BF95CWzH.d.ts} +2 -2
- package/dist/{engine-BpGGw6cd.d.ts.map → engine-BF95CWzH.d.ts.map} +1 -1
- package/dist/{eval-campaign-BUMIxKTE.js → eval-campaign-DoX5vz3E.js} +2 -2
- package/dist/{eval-campaign-BUMIxKTE.js.map → eval-campaign-DoX5vz3E.js.map} +1 -1
- package/dist/{evaluation-CBdQt_kr.d.ts → evaluation-D1PGzkPR.d.ts} +3 -3
- package/dist/{evaluation-CBdQt_kr.d.ts.map → evaluation-D1PGzkPR.d.ts.map} +1 -1
- package/dist/evaluation-wdGyZ7jV.js +147 -0
- package/dist/evaluation-wdGyZ7jV.js.map +1 -0
- package/dist/evaluation.d.ts +1 -1
- package/dist/evaluation.js +1 -136
- package/dist/{exact-types-BYFmYq44.d.ts → exact-types-CVn1QxFO.d.ts} +2 -2
- package/dist/{exact-types-BYFmYq44.d.ts.map → exact-types-CVn1QxFO.d.ts.map} +1 -1
- package/dist/experiment/index.d.ts +3 -3
- package/dist/experiment/index.js +2 -2
- package/dist/{external-optimizer-process-wOLfzsCC.js → external-optimizer-process-E5yY2SYS.js} +2 -2
- package/dist/{external-optimizer-process-wOLfzsCC.js.map → external-optimizer-process-E5yY2SYS.js.map} +1 -1
- package/dist/{external-optimizer-subprocess-D670Oxf2.js → external-optimizer-subprocess-DA02c_L6.js} +12 -19
- package/dist/{external-optimizer-subprocess-D670Oxf2.js.map → external-optimizer-subprocess-DA02c_L6.js.map} +1 -1
- package/dist/{feedback-trajectory-DhB9WzYg.d.ts → feedback-trajectory-BZhVCWuz.d.ts} +3 -3
- package/dist/{feedback-trajectory-DhB9WzYg.d.ts.map → feedback-trajectory-BZhVCWuz.d.ts.map} +1 -1
- package/dist/finding-composition-8n0fFgVI.js +131 -0
- package/dist/finding-composition-8n0fFgVI.js.map +1 -0
- package/dist/{findings-store-CvNo712L.js → findings-store-DRiFdoid.js} +4 -4
- package/dist/{findings-store-CvNo712L.js.map → findings-store-DRiFdoid.js.map} +1 -1
- package/dist/{index-CVFStrQD.d.ts → index-BJZ7s-sv.d.ts} +7 -58
- package/dist/index-BJZ7s-sv.d.ts.map +1 -0
- package/dist/{index-Q6DvdcwI.d.ts → index-BvX3CFNS.d.ts} +6 -6
- package/dist/{index-Q6DvdcwI.d.ts.map → index-BvX3CFNS.d.ts.map} +1 -1
- package/dist/index-DWd1gUwR.d.ts +1086 -0
- package/dist/index-DWd1gUwR.d.ts.map +1 -0
- package/dist/{index-DAXfpVf7.d.ts → index-y7i8l6Y_.d.ts} +8 -8
- package/dist/{index-DAXfpVf7.d.ts.map → index-y7i8l6Y_.d.ts.map} +1 -1
- package/dist/index.d.ts +18 -17
- package/dist/index.d.ts.map +1 -1
- package/dist/index.js +15 -15
- package/dist/{kind-factory-BhxbtUsQ.js → kind-factory-Dl0PmNQL.js} +2 -2
- package/dist/{kind-factory-BhxbtUsQ.js.map → kind-factory-Dl0PmNQL.js.map} +1 -1
- package/dist/{llm-judge-Cc8vDJh_.js → llm-judge-3uAeRawE.js} +4 -4
- package/dist/{llm-judge-Cc8vDJh_.js.map → llm-judge-3uAeRawE.js.map} +1 -1
- package/dist/{matrix-Cf-AGxGh.d.ts → matrix-BXo3PSqM.d.ts} +2 -2
- package/dist/{matrix-Cf-AGxGh.d.ts.map → matrix-BXo3PSqM.d.ts.map} +1 -1
- package/dist/meta-eval/index.d.ts +3 -638
- package/dist/meta-eval/index.js +3 -193
- package/dist/meta-eval/index.js.map +1 -1
- package/dist/multishot/golden/index.d.ts +1 -1
- package/dist/multishot/index.d.ts +2 -2
- package/dist/openapi.json +1 -1
- package/dist/outcome-observations-FLmRpgYa.js +78 -0
- package/dist/outcome-observations-FLmRpgYa.js.map +1 -0
- package/dist/{paired-arms-Ct4LTKc1.js → paired-arms-B2ttLc0_.js} +108 -31
- package/dist/paired-arms-B2ttLc0_.js.map +1 -0
- package/dist/{power-and-mde-N5nU4_Ly.js → power-and-mde-CQX-0weG.js} +2 -2
- package/dist/{power-and-mde-N5nU4_Ly.js.map → power-and-mde-CQX-0weG.js.map} +1 -1
- package/dist/{prime-protocol-CckHBxGu.d.ts → prime-protocol-BEaqBoQE.d.ts} +2 -2
- package/dist/{prime-protocol-CckHBxGu.d.ts.map → prime-protocol-BEaqBoQE.d.ts.map} +1 -1
- package/dist/{proposal-findings-E1aSC_cU.js → proposal-findings-D0_zkQ2l.js} +2 -2
- package/dist/{proposal-findings-E1aSC_cU.js.map → proposal-findings-D0_zkQ2l.js.map} +1 -1
- package/dist/{registry-B_yG-1jh.d.ts → registry-DDKaUrmM.d.ts} +3 -3
- package/dist/{registry-B_yG-1jh.d.ts.map → registry-DDKaUrmM.d.ts.map} +1 -1
- package/dist/reporting.d.ts +1 -1
- package/dist/reporting.js +3 -3
- package/dist/rl.d.ts +4 -5
- package/dist/rl.d.ts.map +1 -1
- package/dist/rl.js +5 -4
- package/dist/rl.js.map +1 -1
- package/dist/{rubric-predictive-validity-noY7uE03.d.ts → rubric-predictive-validity-Cd5SCtAf.d.ts} +58 -3
- package/dist/rubric-predictive-validity-Cd5SCtAf.d.ts.map +1 -0
- package/dist/{rubric-predictive-validity-D3_oI72N.js → rubric-predictive-validity-CeKW2GQN.js} +3 -77
- package/dist/rubric-predictive-validity-CeKW2GQN.js.map +1 -0
- package/dist/search/index.d.ts +2 -2
- package/dist/{search-kernel-Dloxd6gs.d.ts → search-kernel-C6__V0IS.d.ts} +2 -2
- package/dist/{search-kernel-Dloxd6gs.d.ts.map → search-kernel-C6__V0IS.d.ts.map} +1 -1
- package/dist/{skillopt-optimization-method-DLKqdhcR.js → skillopt-optimization-method-D21dH_Bh.js} +4 -4
- package/dist/{skillopt-optimization-method-DLKqdhcR.js.map → skillopt-optimization-method-D21dH_Bh.js.map} +1 -1
- package/dist/{statistical-heldout-3z3I-kJN.d.ts → statistical-heldout-pD1Yah-0.d.ts} +58 -8
- package/dist/statistical-heldout-pD1Yah-0.d.ts.map +1 -0
- package/dist/{store-otlp-dyWE3wAG.js → store-otlp-BVZfR168.js} +2 -2
- package/dist/{store-otlp-dyWE3wAG.js.map → store-otlp-BVZfR168.js.map} +1 -1
- package/dist/{store-tool-spans-DIUgZbzu.js → store-tool-spans-CjYiF4fF.js} +3 -3
- package/dist/{store-tool-spans-DIUgZbzu.js.map → store-tool-spans-CjYiF4fF.js.map} +1 -1
- package/dist/{store-tool-spans-DPwXHQkH.d.ts → store-tool-spans-DfnwWMFR.d.ts} +4 -4
- package/dist/{store-tool-spans-DPwXHQkH.d.ts.map → store-tool-spans-DfnwWMFR.d.ts.map} +1 -1
- package/dist/{summary-report-BsXJoYX5.js → summary-report-DHDWxe6p.js} +3 -3
- package/dist/{summary-report-BsXJoYX5.js.map → summary-report-DHDWxe6p.js.map} +1 -1
- package/dist/systemone-D3MrW63C.d.ts +400 -0
- package/dist/systemone-D3MrW63C.d.ts.map +1 -0
- package/dist/{systemone-protocol-DKoi1NdK.d.ts → systemone-protocol-Bl2EI2KQ.d.ts} +4 -2
- package/dist/{systemone-protocol-DKoi1NdK.d.ts.map → systemone-protocol-Bl2EI2KQ.d.ts.map} +1 -1
- package/dist/systemone-protocol.d.ts +2 -2
- package/dist/systemone-protocol.js +7 -7
- package/dist/systemone-protocol.js.map +1 -1
- package/dist/{systemone-review-CUBMuTLi.d.ts → systemone-review-C8DRiT_-.d.ts} +3 -3
- package/dist/{systemone-review-CUBMuTLi.d.ts.map → systemone-review-C8DRiT_-.d.ts.map} +1 -1
- package/dist/{systemone-review-DzdOJ8I6.js → systemone-review-DFGtIUcf.js} +2 -2
- package/dist/{systemone-review-DzdOJ8I6.js.map → systemone-review-DFGtIUcf.js.map} +1 -1
- package/dist/systemone.d.ts +5 -5
- package/dist/systemone.js +5 -343
- package/dist/{tool-groups-LISbBWCg.d.ts → tool-groups-DK1bu5TD.d.ts} +3 -3
- package/dist/tool-groups-DK1bu5TD.d.ts.map +1 -0
- package/dist/{trace-questions-C42fYEE4.d.ts → trace-questions-D_OKYZ6u.d.ts} +3 -3
- package/dist/{trace-questions-C42fYEE4.d.ts.map → trace-questions-D_OKYZ6u.d.ts.map} +1 -1
- package/dist/trace-repair/index.d.ts +1 -1
- package/dist/traces.d.ts +3 -3
- package/dist/traces.js +3 -3
- package/dist/{types-DqFMNTgK.js → types-CnWmj8NW.js} +19 -6
- package/dist/types-CnWmj8NW.js.map +1 -0
- package/dist/{types-CHfn1zg5.d.ts → types-DDKfH-N6.d.ts} +2 -2
- package/dist/{types-CHfn1zg5.d.ts.map → types-DDKfH-N6.d.ts.map} +1 -1
- package/dist/{types-8yQzNm8u.d.ts → types-DGImvHIr.d.ts} +150 -4
- package/dist/types-DGImvHIr.d.ts.map +1 -0
- package/dist/{types-D8GXK9Z5.d.ts → types-IERw_HxR.d.ts} +2 -2
- package/dist/{types-D8GXK9Z5.d.ts.map → types-IERw_HxR.d.ts.map} +1 -1
- package/dist/wire/index.d.ts +1 -1
- package/docs/behavioral-trace-questions.md +99 -0
- package/docs/experiment.md +27 -0
- package/docs/finding-layers.md +54 -1
- package/docs/systemone.md +107 -0
- package/package.json +4 -4
- package/dist/benchmark-command-Dx0E4a6z.js.map +0 -1
- package/dist/completion-verifier-Cigap8CG.js.map +0 -1
- package/dist/default-registry-Cwg_QN3d.d.ts +0 -70
- package/dist/default-registry-Cwg_QN3d.d.ts.map +0 -1
- package/dist/default-registry-ovrIBGmH.js.map +0 -1
- package/dist/evaluation.js.map +0 -1
- package/dist/index-CVFStrQD.d.ts.map +0 -1
- package/dist/judge-gate-Dc_mXQsL.d.ts +0 -453
- package/dist/judge-gate-Dc_mXQsL.d.ts.map +0 -1
- package/dist/meta-eval/index.d.ts.map +0 -1
- package/dist/outcome-store-CNt4iZ67.d.ts +0 -58
- package/dist/outcome-store-CNt4iZ67.d.ts.map +0 -1
- package/dist/paired-arms-Ct4LTKc1.js.map +0 -1
- package/dist/rubric-predictive-validity-D3_oI72N.js.map +0 -1
- package/dist/rubric-predictive-validity-noY7uE03.d.ts.map +0 -1
- package/dist/statistical-heldout-3z3I-kJN.d.ts.map +0 -1
- package/dist/systemone-DVA7t76V.d.ts +0 -135
- package/dist/systemone-DVA7t76V.d.ts.map +0 -1
- package/dist/systemone.js.map +0 -1
- package/dist/tool-groups-LISbBWCg.d.ts.map +0 -1
- package/dist/types-8yQzNm8u.d.ts.map +0 -1
- package/dist/types-DqFMNTgK.js.map +0 -1
package/dist/rl.d.ts
CHANGED
|
@@ -1,18 +1,17 @@
|
|
|
1
1
|
import { _ as Span } from "./schema-D6EVKzTB.js";
|
|
2
2
|
import { a as RunRecord, c as RunSplitTag } from "./run-record-C3awiXPf.js";
|
|
3
3
|
import { a as PairedPromotionDecision, u as PairedBootstrapResult } from "./paired-promotion-decision-CQMJ5xGN.js";
|
|
4
|
-
import { a as CampaignResult } from "./types-
|
|
4
|
+
import { a as CampaignResult } from "./types-DDKfH-N6.js";
|
|
5
5
|
import { a as MintedRolloutLine, b as RolloutSplit, t as ChatMessage } from "./schema-C-ziyJD7.js";
|
|
6
6
|
import { t as MintRolloutOptions } from "./mint-CkzpVPdz.js";
|
|
7
7
|
import { a as detectRewardHacking, c as VerifiableRewardSource, i as RewardHackingSignal, l as extractVerifiableReward, n as RewardHackingFinding, o as VerifiableReward, r as RewardHackingReport, s as VerifiableRewardExtractionOptions, t as DetectRewardHackingInput, u as extractVerifiableRewardsFromRecords } from "./reward-hacking-WTxUbtbb.js";
|
|
8
|
-
import {
|
|
9
|
-
import { a as SystemOneReviewReport, t as PreparedSystemOneReview } from "./systemone-review-
|
|
10
|
-
import { n as TraceQuestionOutcome, r as TraceQuestionSpec } from "./trace-questions-
|
|
8
|
+
import { a as RubricPredictiveValidityReport, d as FileSystemOutcomeStore, f as FileSystemOutcomeStoreOptions, h as OutcomeStore, p as InMemoryOutcomeStore, t as OutcomeMetricSpec, u as DeploymentOutcome } from "./rubric-predictive-validity-Cd5SCtAf.js";
|
|
9
|
+
import { a as SystemOneReviewReport, t as PreparedSystemOneReview } from "./systemone-review-C8DRiT_-.js";
|
|
10
|
+
import { n as TraceQuestionOutcome, r as TraceQuestionSpec } from "./trace-questions-D_OKYZ6u.js";
|
|
11
11
|
import { t as InterimReleaseConfidence } from "./sequential-KwGspfeR.js";
|
|
12
12
|
import { t as AdversarialMutation } from "./adversarial-smnADNFS.js";
|
|
13
13
|
import { a as SteeringChange, f as EvalCampaignOptions, h as runEvalCampaign, i as Researcher, n as ExperimentResult, p as EvalCampaignResult, r as FailureMode, t as ExperimentPlan } from "./researcher-H0p44OYC.js";
|
|
14
14
|
import { o as TrainingExportOptions } from "./exporters-Dk3oEbK_.js";
|
|
15
|
-
import { a as RubricPredictiveValidityReport, t as OutcomeMetricSpec } from "./rubric-predictive-validity-noY7uE03.js";
|
|
16
15
|
//#region src/rl/active-curriculum.d.ts
|
|
17
16
|
/**
|
|
18
17
|
* Adaptive curriculum: active scenario selection by variance.
|
package/dist/rl.d.ts.map
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"rl.d.ts","names":[],"sources":["../src/rl/active-curriculum.ts","../src/rl/adaptation-eval.ts","../src/rl/claim-integrity.ts","../src/rl/compute-curves.ts","../src/rl/contamination.ts","../src/rl/rollout-input.ts","../src/rl/preferences.ts","../src/rl/process-reward.ts","../src/rl/exporters.ts","../src/rl/dataset.ts","../src/rl/corpus.ts","../src/rl/off-policy.ts","../src/rl/predictive-validity-researcher.ts","../src/rl/rl-campaign.ts","../src/rl/run-record-adapters.ts","../src/rl/sim-fidelity.ts","../src/rl/tournament.ts","../src/rl/verified-findings-dataset.ts"],"mappings":"
|
|
1
|
+
{"version":3,"file":"rl.d.ts","names":[],"sources":["../src/rl/active-curriculum.ts","../src/rl/adaptation-eval.ts","../src/rl/claim-integrity.ts","../src/rl/compute-curves.ts","../src/rl/contamination.ts","../src/rl/rollout-input.ts","../src/rl/preferences.ts","../src/rl/process-reward.ts","../src/rl/exporters.ts","../src/rl/dataset.ts","../src/rl/corpus.ts","../src/rl/off-policy.ts","../src/rl/predictive-validity-researcher.ts","../src/rl/rl-campaign.ts","../src/rl/run-record-adapters.ts","../src/rl/sim-fidelity.ts","../src/rl/tournament.ts","../src/rl/verified-findings-dataset.ts"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;UAgBiB;EACf;EACA;;EAEA;;UAGe;EACf;EACA;;EAEA;;EAEA;;UAGe;;EAEf;;;;;;EAMA;;;;;EAKA;;;;iBChCe,iBAAiB;;;;;;EAMhC,IAAI;IAAQ,UAAU;IAAG;IAAW;MAAgB;;iBAGrC,0BAA0B;EACzC,WAAW;;EAEX;;EAEA;EACA,QAAQ,iBAAiB;;EAEzB;;iBAGe;EACf;EACA;EACA;EACA;EACA;;EAEA,aAAa;IAAQ;IAAoB;IAAmB;IAAgB;;;iBAG7D;EACf,QAAQ;;;;;EAKR;;;;;EAKA;;wBAGoB,mBAAmB;EAAY;GACnD,MAAM,0BAA0B,KAC/B,QAAQ;iBAmEM;EACf,MAAM;IACJ;;IAEA,OAAO;;;EAGT,WAAW;;EAEX,cAAc;EACd,cAAc;;EAEd;;EAEA;;EAEA;;;;;;;;;wBAUc,wBACd,GAAG,iBACH,GAAG,iBACH;EACE;EACA;EACA;;EAEA;IAED;;wBA0Ea,WAAW,OAAO,iBAAiB;;;KC5LvC;KACA;;UAGK;WACN;WACA,IAAI;WACJ,OAAO,sBAAsB;;UAGvB;;WAEN;WACA,eAAe;;WAEf;;WAEA;;WAEA;;UAGM;WACN;WACA;;;;;;;UAQM;;WAEN;;WAEA;;WAEA;;WAEA;;WAEA,SAAS,SAAS;;;;;WAKlB,aAAa,SAAS;;WAEtB,mBAAmB;;WAEnB,oBAAoB;;WAEpB,kBAAkB;;WAElB,mBAAmB;;;UAIb;;WAEN;;WAEA;WACA;WACA;WACA;;WAEA,YAAY,SAAS;;WAErB;;WAEA;;WAEA;;KAGC;KACA;UAYK;WACN;WACA;WACA;;UAGM;WACN,IAAI;WACJ,MAAM;;WAEN;WACA;WACA,mBAAmB;;WAEnB;;;wBA2BK,UAAU,MAAM,WAAW,OAAO,KAAK;;;;;;wBAyDvC,aACd,OAAO,gBACP,OAAO,0BACN;UAsCc;;WAEN;;;;;;;;;wBAUK,cACd,OAAO,gBACP,qBAAqB,kBACrB,OAAO,yBACP,UAAS,iBACR;;;;;;;;;;;qBAkFU;WAEL;WAEF;;WAGE;WAEF;;WAGE;WACK;;WAGL;WACK;;;wBAKG,sBAAsB,kCAAkC;;UAqCvD;WACN;WACA;;WAEA;;;;;;;wBAQK,kBACd,iBAAiB,mBACjB;EAAW;EAA6B;IACvC;cA6DG;KACM,yBAAyB;;wBAkBrB,2BAA2B,OAAO,gBAAgB,OAAO;EAa7D;;;EACA;;;UAIK;WACN,SAAS;WACT;WACA;WACA;;;wBAyBK,wBAAwB,eAAe;;;;;wBA8BvC,cACd,OAAO,gBACP,QAAQ;WAAsC;UAC9C;EAAW;EAAgB;IAC1B;;;;;cA8CG;WACJ;aACE;aACA;;WAGF;aACE;aACA;;WAGF;aACE;aACA;;WAEF;aACE;aACA;;;KAKQ,0CAA0C;UAErC;WACN;;;;;WAKA;;WAEA;WACA;;;wBAIK,8BACd,OAAO,gBACP,OAAO,aACP,SAAS,iCACR,wBAAwB;;wBA+BX,gBACd,QAAQ;WAAmC;UAC3C;EAAW;IACV;;;;;;wBAqCa,6BAA6B,OAAO,iBAAiB;;wBAmBrD,cACd,mBAAmB;WAAoC;UACvD;EAAW;IACV;UAwCc;;WAEN,sBAAsB;;UAGhB;WACN;WACA;;WAEA;WACA,kBAAkB;;WAElB;;WAEA;;;wBAIK,qBAAqB,SAAQ;;wBAS7B,sBACd,kBAAkB,mBAClB,SAAQ,uBACP;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;UC5zBc;;EAEf;;EAEA;;EAEA,OAAO;;UAGQ;EACf;EACA;EACA;;EAEA;;EAEA;;EAEA,UAAU;;UAGK;EACf;EACA,QAAQ;;EAER;;EAEA,MAAM;;UAGS;EACf;EACA,SAAS;;;;;EAKT,cAAc,QAAQ,uBAAuB;IAC3C;IACA;IACA;IACA,UAAU;;;UAuBG,sBAAsB;;EAErC;;EAEA,SAAS,sBAAsB,QAAQ;;EAEvC,UAAU,SAAS,MAAM;;UAGV,qBAAqB;EACpC,MAAM;EACN;EACA;EACA;;EAEA;;;wBAIoB,QAAQ,GAAG,MAAM,sBAAsB,KAAK,QAAQ,qBAAqB;UAqB9E,uBAAuB;EACtC;EACA,SAAS,sBAAsB,QAAQ;;EAEvC,YAAY,SAAS;;UAGN,sBAAsB;;EAErC;;EAEA;;EAEA,WAAW;;EAEX,gBAAgB;;EAEhB,UAAU;;;;;;;;UA2CK;EACf;EACA;EACA;EACA;;wBAGc,wBAAwB,QAAQ,qBAAqB;;;;;;;;;;;;KC7LzD;UAOK,qBAAqB;EACpC,MAAM;;EAEN,QAAQ,UAAU,MAAM,QAAQ,KAAK;;EAErC;;UAGe,wBAAwB;;EAEvC,aAAa,GAAG;;EAEhB,WAAW;;;;;EAKX,YAAY;EACZ,eAAe,qBAAqB;;;;;;EAMpC,UAAU,GAAG,MAAM;;UAGJ;;EAEf;;EAEA;;;;;EAKA;;UAGe;EACf,aAAa;IACX;IACA;IACA;IACA;;;EAGF;IAAc;IAAW;;;EAEzB;EACA;EACA;EACA;;EAEA;;EAEA;;wBAGoB,sBAAsB,GAC1C,OAAO,wBAAwB,IAC/B,OAAM,4BACL,QAAQ;;;;;;;;;;;;;UC5CM;;;;;;;;;;;EAWf,OAAO;;;;YCKG;iBAKK;;EAEf;;EAEA;EACA;;EAEA;EACA;;EAEA;;;;;;EAMA;IAAW;IAAgB;;;EAE3B;;;;;EAKA;IACE;IACA;IACA;IACA;IACA;IACA;;;iBAIa;EACf,WAAW;;;;;EAKX;;;;;;EAMA,QAAQ;;EAER;;iBAGe;EACf,OAAO;;EAEP;;EAEA;;EAEA;;EAEA,UAAU;;;;;;EAMV;;;;;;;;;;;;;;;;;;;;wBAuCc,mBACd,OAAO,qBACP,OAAM,4BACL;;;;;;;;;;;;;;;;;;wBAqMmB,YACpB,SAAS,oBACT,SAAS,YACT,SAAS,qBACR,QAAQ;EAAQ;EAAgB;EAAgB;;;;;;;;;;;wBAkCnC,kBACd,SAAS,oBACT,SAAS,qBACR;EAAQ;EAAoB;EAAqB;EAAuB;;;;UChX1D;;EAEf;EACA;;EAEA;;EAEA,MAAM;;EAEN;;EAEA;;;;;EAKA;;EAEA;;EAEA;;UAGe;;EAEf,WAAW;;EAEX,MAAM,MAAM,OAAO,QAAQ,KAAK;;UAGjB;;;;;;EAMf,SAAS;;EAET,aAAa,MAAM;;UAuCJ;EACf;EACA;EACA;;EAEA;;EAEA;;EAEA;EACA;;UAkDe;;EAEf;EACA;;EAEA;EACA;;EAEA;EACA;;EAEA;EACA;;;;;;;;;;;;;;;;;wBAkBc,iBACd,kBAAkB,YAAY,eAC9B;EAAQ;EAAoB;IAC3B;;;UCrJc;;EAEf,WAAW,2BAA2B;;EAEtC,eAAe,2BAA2B;;UAG3B;EACf;EACA;EACA;;EAEA;;EAEA,OAAO;;;KAIG,iBAAiB;;;;;;;;;;;;;;;;wBAwBP,UACpB,SAAS,oBACT,SAAS,YACT,SAAS,iBACR,QAAQ;;KA+CC,+BAA+B;UAE1B,oBACP,KAAK;;EAEb,WAAW,2BAA2B;;EAEtC,eAAe,2BAA2B;;UAG3B;EACf;EACA;EACA;;EAEA;EACA,OAAO;;;;;;;;;;;;;;;;;;;wBAoBa,WACpB,OAAO,qBACP,SAAS,cACR,QAAQ;UA+DM,mBAAmB;;;;;EAKlC,WAAW,2BAA2B;;EAEtC,eAAe,2BAA2B;;EAE1C,YAAY,MAAM;;EAElB,WAAW,MAAM;;UAGF;;EAEf,UAAU;EACV,OAAO;;;;;;;;;;;;;;;;;;;wBAoBa,qBACpB,OAAO,qBACP,SAAS,aACR,QAAQ;wBAwDK,WAAW,MAAM;UAMhB;;EAEf,WAAW,2BAA2B;;EAEtC,aAAa,eAAe,4BAA4B;;EAExD,YAAY,eAAe,uCAAuC;;UAGnD;EACf;;EAEA;EACA;EACA;EACA;EACA;EACA;EACA;EACA,OAAO;;UAGQ,uBAAuB;;;;;;;;;;EAUtC;;;;;;;;;;;;;;;;;;;;;wBAsBoB,UACpB,SAAS,qBACT,SAAS,YACT,SAAS,iBACR,QAAQ;UAoGM;EACf;EACA;EACA;EACA;EACA;EACA;;;;KCheU;cACN;KACM,wBAAwB;wBAIpB,uBAAuB,iBAAiB;;;UA2BvC;EACf;EACA;;EAEA;;;EAGA;;;EAGA;IAAU,MAAM;IAAY;IAAgB;;EAC5C;EACA;EACA;;EAEA;;EAEA,UAAU;;EAEV;IACE;IACA;IACA;;;UAIa;EACf;EACA;EACA;EACA;EACA;EACA;;UAGe;EACf;;EAEA;;EAEA,QAAQ,OAAO;EACf,QAAQ;;EAER;;EAEA;EACA;EACA;IAAe;IAAe;;EAC9B;;;;;;EAMA;;UAGe,0BAA0B;EACzC,SAAS;EACT,WAAW,QAAQ,OAAO;EAC1B,OAAO;;UAGQ;EACf,UAAU;;EAEV,OAAO;;;;;;;;;;wBA0Da,eACpB,OAAO,qBACP,SAAS,cAAc,YACvB,QAAQ,iBACR;EAAgB,SAAS;EAAoB,SAAS;IACrD,QAAQ;;;;;KC/KC,eAAe;EAAc;EAAiB;;UAEzC;EACf;;;EAGA;EACA;;;wBA8Bc,WAAW,qBAAqB;UAmB/B;;EAEf;;EAEA,SAAS;;EAET;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;UC/Ce;;EAEf;;EAEA;;;;;;EAMA;;;;;;EAMA;;;;;;EAMA;;;;;;;EAOA;;UAGe;;EAEf;;EAEA;;UAGe;;EAEf;;EAEA;;EAEA;;EAEA;;;;;EAKA;;EAEA,qBAAqB;;UAGN;;;;;;EAMf;;EAEA;IAAe;IAAa;;;;;;;;wBAQd,4BACd,cAAc,uBACd,OAAM,mBACL;;;;;;wBA4Ca,kCACd,cAAc,uBACd,OAAM,mBACL;;;;;;;;;;;;;;;;;;;;;wBAwDa,aACd,cAAc,uBACd,OAAM,mBACL;;;iBC7Lc;EACf,UAAU;;EAEV,eAAe;;EAEf;;EAEA;;;;;;EAMA,YAAY,QAAQ,0CAA0C;;;;;;qBAOnD,wCAAwC;mBAClC;UACT;EAER,YAAY,MAAM;EAaZ,gBAAgB,MAAM,cAAc,QAAQ;EA0C5C,cAAc,UAAU,gBAAgB,QAAQ;EA+DhD,YAAY,SAAS,kBAAkB,UAAU,iBAAiB,QAAQ;EAS1E,eAAe,MAAM,iBAAiB,QAAQ;;;;;;EAiD9C,iBAAiB,MAAM,cAAc,QAAQ;;;;;;EAiBnD,UAAU,QAAQ;EAUlB,iBAAiB;;;;iBCxMF,qBAAqB,WAAW,oBAAoB;;EAEnE,UAAU;;EAEV,cAAc;;EAEd,mBAAmB;;EAEnB,eAAe;EACf,0BAA0B;;EAE1B;IACE;IACA;IACA;MAAS;MAAa;;;;;;;;;;;;;IAYtB;;;EAGF;IACE,MAAM;IACN,OAAO;IACP,MAAM;;;;;;;;;;;iBAYO;EACf;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;iBAGe;EACf,UAAU;;EAEV,cAAc;;EAEd,eAAe;IAAQ;IAAe,QAAQ;;;EAE9C,aAAa;;;;;EAKb,mBAAmB;;;;EAInB,eAAe;;EAEf,eAAe;;EAEf,oBAAoB;;EAEpB;IACE,MAAM;IACN,OAAO;IACP,MAAM;;;;;EAKR;;;;EAIA;;wBAGoB,cAAc,GAAG,MAAM,qBAAqB,KAAK,QAAQ;;;UC5I9D;;EAEf;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA,WAAW;;EAEX;;;;;;;;;;;wBAYc,qBACd,UAAU,gBACV,KAAK;EAAmB;IACvB;;;;;;;KCPS,oBAAoB,QAAQ,cAAc;;;wBA0BtC,YAAY,eAAe;UAU1B;;EAEf;;;EAGA;;EAEA;;UAGe;EACf;;EAEA;;EAEA,WAAW;;EAEX;;EAEA;;KAGU;UAEK;EACf,cAAc;;;;EAId;;;EAGA;;;EAGA,SAAS;;UAGM;;EAEf,WAAW;;EAEX;;UAae;;;EAGf;;;EAGA;;UAGe;EACf;EACA;;EAEA;;;EAGA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;UCzHe;;EAEf;;EAEA;;;;;EAKA;;;;;;EAMA;;UAGe;EACf;;EAEA;;EAEA;;EAEA;;EAEA;;UAGe;EACf,SAAS;;EAET;;EAEA;EACA;;;;;;;;;;;wBAYc,gBACd,UAAU,mBACV;EAAQ;EAAoB;EAAwB;IACnD;;;;;;;;;UA+Gc;;EAEf;;EAEA;;;;;;;;UAkCe;EACf,MAAM;IACJ;;IAEA;IACA;;;;;;EAMF;;wBAGc,0BACd,OAAO,iCACN;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;cC/MU;;UAKI;EACf;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;;EAEA;EACA,KAAK;EACL;;UAGe;EACf;EACA;EACA;EACA;EACA;EACA;;UAGe;EACf;EACA,OAAO;;;UAIQ;EACf;EACA;EACA;EACA;EACA;EACA;EACA;EACA,kBAAkB;IAAQ;IAAkB;;;;UAI7B;EACf;EACA;EACA;;UAGe;EACf;;;EAGA;;EAEA;EACA;;;UAIe;EACf;EACA;EACA;EACA;EACA,mBAAmB;EACnB;EACA;IAAU;IAAyB;;;UAKpB;EACf;EACA;EACA;;EAEA;EACA;;KAGU;UAEK;EACf,SAAS;EACT;EACA;EACA;EACA;;UAGe;EACf,eAAe;;EAEf;EACA;EACA;EACA;IACE;IACA;IACA;IACA;IACA;IACA;;EAEF;;IAEE;;IAEA;;IAEA;;IAEA;IACA;;;EAGF;IACE;MAAU;MAAe;;IACzB,OAAO;;EAET;IACE;;IAEA;IACA;IACA;IACA;IACA;IACA;IACA;;;IAGA;IACA,wBAAwB;IACxB;IACA;IACA;IACA;;EAEF,KAAK;EACL;IACE;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;IACA;;;UAIa;EACf;EACA;;;;EAIA;EACA,KAAK,OAAO;EACZ,UAAU;IAAiB;IAAc;IAAoB;;;UAO9C;EACf,WAAW;EACX,OAAO;EACP,OAAO;EACP;EACA;EACA;EACA;EACA;EACA;EACA;EACA,SAAS;EACT;;UA6Oe;EACf;;EAEA;;UAGe;EACf;;EAEA;;EAEA,SAAS,eAAe;;EAExB;EACA;;UAGe;EACf,MAAM;EACN,SAAS;EACT;IACE;IACA;IACA;IACA;IACA,SAAS;MAAiB;MAAoB;MAAsB;;;;;;;;;wBAuExD,4BACd,QAAQ,yBACP"}
|
package/dist/rl.js
CHANGED
|
@@ -1,19 +1,20 @@
|
|
|
1
1
|
import { s as ValidationError } from "./errors-BcOs5MKW.js";
|
|
2
2
|
import { a as hashCanonical, i as compareCodeUnits } from "./canonical-Dq2hOQ8L.js";
|
|
3
3
|
import { o as runTaskScore } from "./run-record-DhEJgoPz.js";
|
|
4
|
+
import { a as wilcoxonSignedRank } from "./paired-arms-B2ttLc0_.js";
|
|
4
5
|
import { a as medianInPlace } from "./internal-BTK91Bqt.js";
|
|
5
|
-
import { a as wilcoxonSignedRank } from "./paired-arms-Ct4LTKc1.js";
|
|
6
6
|
import { i as pairedBootstrap } from "./mean-BbA_JCzW.js";
|
|
7
7
|
import { t as decidePairedPromotion } from "./paired-promotion-decision-D0SHb-ZL.js";
|
|
8
8
|
import { n as campaignCellToRunRecord } from "./run-record-CysKLB93.js";
|
|
9
9
|
import { n as extractVerifiableReward, r as extractVerifiableRewardsFromRecords, t as detectRewardHacking } from "./reward-hacking-ByuETaIb.js";
|
|
10
|
-
import { t as runEvalCampaign } from "./eval-campaign-
|
|
10
|
+
import { t as runEvalCampaign } from "./eval-campaign-DoX5vz3E.js";
|
|
11
11
|
import { s as assertRewardGate } from "./schema-CvhfLjtZ.js";
|
|
12
12
|
import { n as isTrainingLineEligible, o as toSftRows, t as isSplitEligible } from "./exporters-DWvM16LE.js";
|
|
13
13
|
import { t as mintRolloutRows } from "./mint-CbdwjTPW.js";
|
|
14
14
|
import { t as evaluateInterimReleaseConfidence } from "./sequential-BVv1sjsw.js";
|
|
15
|
-
import {
|
|
16
|
-
import { n as
|
|
15
|
+
import { o as validateOutcomeMetricSpecifications, t as assertUniqueObservationIds } from "./outcome-observations-FLmRpgYa.js";
|
|
16
|
+
import { n as prepareSystemOneReview } from "./systemone-review-DFGtIUcf.js";
|
|
17
|
+
import { t as rubricPredictiveValidity } from "./rubric-predictive-validity-CeKW2GQN.js";
|
|
17
18
|
import { n as InMemoryOutcomeStore, t as FileSystemOutcomeStore } from "./outcome-store-DUcMj-3D.js";
|
|
18
19
|
import { createHash } from "node:crypto";
|
|
19
20
|
import { existsSync, readFileSync } from "node:fs";
|