dpsk-v4-flash-data / README.md
simonycl's picture
README
66b9e84 verified
|
Raw
History Blame Contribute Delete
1.63 kB
---
tags: [agentic-ptb, dpsk-v4-flash, training-data]
---
# dpsk-v4-flash-data
Training data built by the AgentPTB arm for cell **`dpsk-v4-flash`** — pi / DeepSeek v4-flash @ effort `thinking`.
This is the corpus the arm *itself* assembled during its 100-hour run: what it downloaded,
filtered, rewrote and mixed. It is the input side of the checkpoints published as
`agentic-ptb/dpsk-v4-flash.h*`, and the companion to the run record in `agentic-ptb/dpsk-v4-flash-record`.
| field | value |
|---|---|
| plot cell | `dpsk-v4-flash` |
| driver | pi / DeepSeek v4-flash |
| reasoning effort | `thinking` |
| total size | 18.31 GB |
| path in run | `/mnt/pvc/users/simon/agentptb/runs/c/workspace/sft_data` |
## Largest components
| directory | size |
|---|---|
| `raw` | 2.82 GB |
| `ds_main2` | 2.24 GB |
| `ds_deep2` | 1.65 GB |
| `ds_deep` | 1.23 GB |
| `ds_main` | 1.17 GB |
| `ds_final` | 0.98 GB |
| `ds_smoke` | 0.04 GB |
## A note on apparent secrets
Automated scanning flags credential-shaped strings (`AKIA…`, `sk-…`, `AIza…`) in these
corpora. They were checked: they are **synthetic fixtures inside the training tasks
themselves** — secret-scanning and config-hygiene exercises whose task text embeds fake keys
(`"secret_type": "api_key"`, `"matched_pattern": …`). No credential belonging to this project
appears in the data.
## Related
- `agentic-ptb/dpsk-v4-flash-record` — driver trajectory, harness, configs, evals for this cell
- `agentic-ptb/dpsk-v4-flash.h*` — the checkpoints trained on this data
- `agentic-ptb/INDEX` — manifest joining every checkpoint back to the sweep figures