blakeslee-research/nucleic-purpose-classifier
3
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge nucleic/plucky-north-vole-sdna into dev

Browse Source
This commit is contained in:
abkslm
2026-08-01 06:29:50 -07:00
parent 2ed9d28c28
commit bb4dff9b56
5 changed files with 161 additions and 112 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+6 -6
View File
@@ -65,9 +65,9 @@ Nucleic managed container.
The SWE-chat source is not downloaded by this repository. After accepting the dataset's
Hugging Face conditions, place a **pinned** Parquet snapshot below the ignored
`.artifacts/swe-chat/raw/` directory, record its immutable revision, then run the
streaming extractor. It reads only the needed columns, takes the first three qualifying
human prompts per session, and writes the first prompt plus hashes for the two context
turns. Do not use `main` as a revision.
streaming extractor. It reads only the needed columns, takes the first qualifying human
prompt plus its conversational agent response, and writes the prompt plus a response hash.
Do not use `main` as a revision.
```bash
ml/purpose-classifier/.venv/bin/pip install -r \
@@ -77,11 +77,11 @@ ml/purpose-classifier/.venv/bin/python \
--revision <accepted-immutable-hf-revision>
```
The export and manifest remain ignored because candidate JSONL temporarily contains all
three messages. Run the one-record schema/availability canary before the 100-session dry
The export and manifest remain ignored because candidate JSONL temporarily contains the
prompt and agent response. Run the one-record schema/availability canary before the 100-session dry
run; both use Luna through subscription-backed `codex exec`, not an API key. The labeler
writes only the first message to canonical source JSONL; state and audit sidecars retain
the other turns solely as hashes and source IDs.
the response solely as a hash and source ID.
```bash
ml/purpose-classifier/.venv/bin/python \