model card for k3-sec-8b-v4 (branch main)

README.md

@@ -0,0 +1,86 @@
+---
+license: apache-2.0
+base_model: Qwen/Qwen3-8B
+tags: [cybersecurity, agentic, red-team, blue-team, distillation, glm-5.2, kimi-k3]
+---
+
+# k3-sec-8b (v4)
+
+An 8B cybersecurity agent (offense + defense) fine-tuned from `Qwen/Qwen3-8B` on
+**699 verified agentic + Q&A traces**, distilled from GLM-5.2 (round 4) and Kimi K3/K2.6
+(rounds 1–3). Trained to operate an autonomous security harness — plan, run bash, read
+observations, write files, finish — not just answer security questions.
+
+Pipeline per iteration: failure analysis on eval traces → parameterized, decontaminated
+seed factory → teacher best-of-3 rejection-sampled agentic transcripts (judged +
+artifact-checked) → full-FT SFT → multi-run attack/defend eval → next round.
+
+## Version 4 highlights
+
+- **Data**: 699 unique rows (605 agentic harness transcripts + 94 Q&A). Round 4 added
+  564 GLM-5.2 traces targeting the five v3 tasks that never passed, generated from 580
+  parameterized seeds with decontamination-by-construction (every eval-graded string is
+  blacklisted and asserted absent). Best-of-3 rejection with a glm-4.7-flash judge
+  (kept 97%). Note: v3's advertised 157 rows contained only 135 unique after legacy
+  merge duplicates; v4 is a genuine 5.2× data increase.
+- **Training**: full FT bf16, 2 epochs, lr 1e-5 cosine, eff. batch 32, seq 8192,
+  adamw_8bit, 5.37M tokens, ~42 min on 1× A100-80GB. **train_loss 0.977 · token-acc
+  80.0%** (v3: 1.395 / 69.1%). Ships with Qwen3 YaRN `rope_scaling` for 131072-token
+  serving.
+- **Eval** (fixed 10-task synthetic attack/defend lab, agentic harness, 3 runs):
+  **mean 6.67/10** (runs: 9, 6, 5) vs v3 mean 4.0/10 (4 runs). Peak run met the
+  promotion gate (≥8/10, both sides ≥4/5) once in three.
+
+### Per-task pass rates (P across runs)
+
+| Task | v3 (4 runs) | v4 (3 runs) |
+|---|---|---|
+| atk-sqli | 0/4 | 0/3 |
+| atk-hash | 4/4 | 3/3 |
+| atk-re | 3/4 | 2/3 |
+| atk-dns | 0/4 | 1/3 |
+| atk-jwt | 0/4 | **3/3** |
+| def-bruteforce | 3/4 | 3/3 |
+| def-webshell | 2/4 | 2/3 |
+| def-harden | 4/4 | 3/3 |
+| def-detect-dns | 0/4 | 1/3 |
+| def-ioc | 0/4 | 2/3 |
+
+Run-to-run variance is significant at temperature 0.7; single-run scores are not
+meaningful for this suite. Known v4 gap: `atk-sqli` — the model prefers to start the
+staged vulnerable app and fuzz it over HTTP instead of reading the offline artifacts
+(trace-verified behavioral prior, targeted in round 5).
+
+## Usage
+
+vLLM, short-task/eval serving (disable static YaRN):
+
+```bash
+python3 -m vllm.entrypoints.openai.api_server \
+  --model bebrws/k3-sec-8b --revision v4 \
+  --port 8000 --hf-overrides '{"rope_scaling":null}' --max-model-len 32768
+```
+
+Long-context serving: omit `--hf-overrides` and set `--max-model-len 131072`.
+Recommended sampling for agentic loops (non-thinking): `temperature=0.7 top_p=0.8
+top_k=20 min_p=0`, `chat_template_kwargs.enable_thinking=false`, per-step completion
+cap ~4096 tokens.
+
+## Intended use & limitations
+
+Defensive/offensive **security research artifact**, evaluated on a small synthetic lab.
+Not for: real intrusion activity, exploit weaponization, unsupervised security
+decisions, or non-security tasks. Outputs require qualified human review. Attack-side
+competence is deliberately scoped to CTF/lab-grade tasks.
+
+## Version history (8B lineage)
+
+| Version | Data | Eval mean | Notes |
+|---|---|---|---|
+| v1 | 135 traces | 4/10 single | first 8B run |
+| v2 | 149 traces | 6/10 single | failure-targeted r2 |
+| v3 | 157 (135 unique) | 4.0/10 (4 runs) | parser-fixed harness baseline |
+| **v4** | **699** | **6.67/10 (3 runs)** | GLM-5.2 scale-up, jwt fixed |
+
+Weights are Apache-2.0 per the Qwen3 base; training traces were generated by GLM-5.2
+and Kimi teachers and filtered by automated judging.