Back to corpus
experimentexperiment writeup candidatescore 24
Experiment Index: cc-semantic-language Phase 9
**Version**: 1.0.0 **Status**: Active **Schema Version**: 1.0.0 **Last Modified**: 2024-12-31 **Mode**: Simulated (BATCH-001) — See RUN_MANIFEST.md
Full HTML reader
Read the full artifact
Extracted abstract or opening context
**Version**: 1.0.0 **Status**: Active **Schema Version**: 1.0.0 **Last Modified**: 2024-12-31 **Mode**: Simulated (BATCH-001) — See RUN_MANIFEST.md
This document is the table of contents for reproducibility. It enumerates every experiment run and maps to file paths for inputs, results, and logs. If someone else runs this repository, this index makes it impossible to get lost.
| Field | Description | |-------|-------------| | **Experiment ID** | Unique identifier (EXP-XXX) | | **Stress Profile** | Name of stress profile used | | **Seed** | Random seed for PRNG | | **Replay Count** | Number of rehearsal iterations | | **Target Stage** | Starting lifecycle stage filter | | **Dataset Slice** | Manifest reference | | **Input Path** | JSONL input file relative to `repro/phase9_bundle_v1/` | | **Result Path** | JSONL result file relative to `repro/phase9_bundle_v1/` | | **Event Log** | Event log snapshot hash or path | | **Status** | Pending / Complete / Failed |
| Field | Value | |-------|-------| | Experiment ID | EXP-001 | | Stress Profile | baseline_v1 | | Seed | 42 | | Replay Count | 100 | | Target Stage | proto | | Dataset Slice | publication_slice_v1 | | Input Path | runs/baseline_42_input.jsonl | | Result Path | runs/baseline_42_result.jsonl | | Event Log | logs/baseline_42_events.hash | | Status | Pending |
| Field | Value | |-------|-------| | Experiment ID | EXP-002 | | Stress Profile | baseline_v1 | | Seed | 100 | | Replay Count | 100 | | Target Stage | proto | | Dataset Slice | publication_slice_v1 | | Input Path | runs/baseline_100_input.jsonl | | Result Path | runs/baseline_100_result.jsonl | | Event Log | logs/baseline_100_events.hash | | Status | Pending |
Promotion decision
What has to happen next
Attach run IDs, datasets, metrics, and reproduction commands.
Why this is not always a full paper yet
Corpus pages are public-safe readers for discovered workspace artifacts. They are not automatically final papers. A corpus item becomes a polished paper only after the editable source, evidence checkpoints, references, figures, render path, and release status are attached through the paper schema.