EvalSuiteResultV1
Defined in: core/src/types/eval.ts:739
Result of one suite.
Properties
dataset
dataset:
EvalDatasetRefV1
Defined in: core/src/types/eval.ts:750
Dataset the suite ran on
findings
findings:
EvalFindingV1[]
Defined in: core/src/types/eval.ts:758
Dataset findings
judges?
optionaljudges?:EvalJudgeRefV1[]
Defined in: core/src/types/eval.ts:752
Judge rubrics the suite used, sorted by rubric id
models
models:
EvalModelResultV1[]
Defined in: core/src/types/eval.ts:760
Model entries in suite order
name
name:
string
Defined in: core/src/types/eval.ts:741
Suite name
passed
passed:
boolean
Defined in: core/src/types/eval.ts:754
Whether every threshold held
prompt
prompt:
EvalPromptRefV1
Defined in: core/src/types/eval.ts:748
Prompt under test
suiteHash
suiteHash:
string
Defined in: core/src/types/eval.ts:746
SHA-256 of the suite file bytes.
Pattern
^sha256-[0-9a-f]{64}$
thresholds
thresholds:
EvalThresholdResultV1[]
Defined in: core/src/types/eval.ts:756
Threshold checks