Prepared for private transfer between the user's evaluation and paper machines. This repository contains existing evaluation data, not new experiments. The two final Qwen pools are: - shared-tb21-native-text-only-xhigh-20260918: 64/89 = 71.9101%.
Dataset Card
Prepared for private transfer between the user's evaluation and paper machines. This repository contains existing evaluation data, not new experiments. The two final Qwen pools are: - shared-tb21-native-text-only-xhigh-20260918: 64/89 = 71.9101%. - shared-tb21-native-residual-xhigh-20260918: 68/89 = 76.4045%. Both live pool directories are under: /inspire/l20d/project/sais-bio/public/hanchen/rm-stack/runs/tb21/jobs/. These are Qwen3.8-27B runs, with native text compaction and step2320 residual weights in the residual arm. They are not the newer GLM pools. All 178 live committed result hashes still match the sealed 2026-09-20 export. case-study.tar.gz is approximately 123 MB. It contains all…
Excerpt from the card by Hanchen Xia.
Details
- Repository
- mocoV3/TB-2-1-Cases
- Publisher
- Hanchen Xia
- Task category
- Not stated by the source
- Tags
- terminal-bench, agent-evaluation, residual-memory
- Size category
- Not stated by the source
- Languages
- en
- Revision
- cd3ae21df55dda7775f374b43cb7e520160b9f23
- Last updated
- 2026-09-24
Files
28 files, 1.7 GB in total.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| case-index.json | Data | 3.4 KB | — |
| case-package-result.json | Data | 3.8 KB | — |
| case-study.tar.gz | Data | 122.6 MB | a802fde07953 |
| full-export-revalidation.json | Data | 71.9 KB | — |
| full-export/archive_manifest.json | Data | 1.4 KB | — |
| full-export/core.files.txt | Data | 609.3 KB | — |
| full-export/core.tar.gz | Data | 43.4 MB | b02c0ace81bd |
| full-export/evaluation_only.files.txt | Data | 166.1 KB | — |
| full-export/evaluation_only.tar.gz | Data | 13.5 MB | 6abd3b9830bd |
| full-export/optional.files.txt | Data | 1.2 MB | — |
| full-export/optional.tar.gz | Data | 300.2 MB | c7bfa0194fa9 |
| full-export/provenance.json | Data | 2.1 KB | — |
| full-export/redaction_report.json | Data | 154.0 KB | — |
| full-export/traces.files.txt | Data | 3.4 MB | — |
| full-export/traces.tar.gz | Data | 1.2 GB | 7c7a99d47d8a |
| full-export/validation_report.json | Data | 103.7 KB | — |
| full-upload-validation.json | Data | 896 B | — |
| package-manifest.json | Data | 778 B | — |
| transfer-validation.json | Data | 294 B | — |
| CASE_STUDY.md | Documentation | 3.4 KB | — |
| README.md | Documentation | 4.5 KB | — |
| full-export/CONFIG_DIFF.md | Documentation | 2.3 KB | — |
| full-export/POOL_IDENTIFICATION.md | Documentation | 2.5 KB | — |
| full-export/README.md | Documentation | 3.2 KB | — |
| full-export/SCHEMA_MAP.md | Documentation | 5.4 KB | — |
| SHA256SUMS | Other | 2.4 KB | — |
| full-export/ARCHIVE_SHA256SUMS | Other | 758 B | — |
| .gitattributes | Repository | 2.5 KB | — |
License and Download
- License
- Not stated by the source
- Access
- No access gate
Released by Hanchen Xia through its official repository on Hugging Face.