Commercial 6,000-Row Multi-Domain Security Suite v4 (Zero Overlap with v1–v3): Need an unseen, contamination-free benchmark to evaluate or fine-tune beyond public splits? Explore the 6,000-row Oxford Digital Enterprise Security Remediation Suite v4 (
1,200Solidity EVM +1,200Rust/Solana Anchor +1,200Python/TS AI Agent & MCP +1,200Go/TS/Python Cloud APIs +1,200K8s/Terraform IaC in Parquet/CSV/JSONL):
- Live Interactive Explorer Space:
Offlin33er/oxford-digital-security-suite-v4· 100-Row Sample:Offlin33er/solidity-exploit-reasoning-v4-sample- Instant Stripe Checkout: $99 Single-Seat Indie → · $599 Team Pro → · $1,499 Enterprise → · Instant USDC / Invoice:
rayxxxnd@oxforddigitals.com
A LoRA adapter for Qwen/Qwen2.5-Coder-7B-Instruct
fine-tuned to produce structured Solidity security-audit reports: given a contract snippet, it emits
## Finding: <type> with functionality analysis, Description, and Recommendation — or a clear
"no vulnerability identified" verdict.
Trained as part of the Solidity Audit Scanner project.
Newer versions available: V2 (canonical 8-class taxonomy, 48.5% type match) and V3 (verified fallback resolution, best detection F1 0.986). This V1 card documents the original run.
| Method | TRL SFTTrainer, QLoRA (NF4, double quant, bf16 compute) |
| LoRA | r=32, alpha=32, dropout 0.05, all attn + MLP projections |
| Data | Offlin33er/solidity-audit-instruct — 2,224 train / 248 held-out, built from Solodit-audited code |
| Epochs | 2 (278 steps, effective batch 16, seq len 2048) |
| Optimizer | paged AdamW 8-bit, LR 2e-4, cosine, warmup 10 |
| Hardware | 1× Nvidia A10G (24 GB), ~2h16m wall |
| Tracking | trackio dashboard |
Training loss (trackio curve): 1.21 @ step 5 → 0.87 @ step 15 → 0.64 @ step 48 → 0.56 @ step 115 → ~0.52 at the end of the cosine tail; token accuracy 0.71 → 0.86.
Stratified sample of the held-out test split — 70 vulnerable / 30 clean rows, seed 42, greedy
decoding, exact training-time system prompt, 512 new tokens. Full per-row records in
eval_results.json.
| Metric | Result |
|---|---|
| Vulnerability detection precision | 1.00 (70/70 flagged rows true) |
| Detection recall | 1.00 (no vulnerable row missed) |
| False-positive rate on clean code | 0.00 (30/30 clean rows correctly cleared) |
| Finding-type match vs reference taxonomy | 0.30 (21/70, normalized substring/keyword match) |
Read this honestly:
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
base = AutoModelForCausalLM.from_pretrained(
"Qwen/Qwen2.5-Coder-7B-Instruct", dtype=torch.bfloat16, device_map="auto"
)
model = PeftModel.from_pretrained(base, "Offlin33er/qwen25-coder-7b-solidity-audit")
tok = AutoTokenizer.from_pretrained("Offlin33er/qwen25-coder-7b-solidity-audit")
messages = [
{"role": "system", "content": "You are a senior smart-contract security auditor. Audit the provided Solidity code. If a vulnerability is present, report it as '## Finding: <type>' followed by the functionality, a Description, and a Recommendation. If none is found, state that no vulnerability was identified and describe what the code does."},
{"role": "user", "content": "Audit the following Solidity code for security vulnerabilities.\n\n```solidity\n<YOUR CODE>\n```"},
]
inputs = tok.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
out = model.generate(**inputs, max_new_tokens=512)
print(tok.decode(out[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))
Defensive security tooling: analyzes code you paste in. Only audit contracts you are authorized to review.
adapter_model.safetensors 162 MBtraining_args.bin 5.9 kBQwen/Qwen2.5-Coder-7B-Instruct