ot-q3_14b-clean

Qwen2.5-7B-Instruct student model fine-tuned by full-parameter SFT (s1 recipe) on Qwen3-14B (OpenThoughts SWAG, V3-attack cleaned) reasoning traces.

This repo is part of a 4-victim study comparing student distillation outcomes when the teacher's reasoning traces are extracted via the V3 attack (-orig) vs. when the V3 attack wrapper is stripped before training (-clean).

How to load a specific epoch

Each epoch_N/ subfolder is a self-contained, loadable HF checkpoint.

from transformers import AutoModelForCausalLM, AutoTokenizer

REPO = "Chia-Mu-Lab/ot-q3_14b-clean"
model = AutoModelForCausalLM.from_pretrained(REPO, subfolder="epoch_5", torch_dtype="bfloat16")
tok = AutoTokenizer.from_pretrained(REPO, subfolder="epoch_5")

Per-epoch evaluation

All numbers are accuracies in percent on the canonical eval suite (GSM8K-MATH500, AIME24, AIME25, JEEBench Math subset strict/partial, LiveCodeBench v5 pass@1). The base row is the Qwen2.5-7B-Instruct starting point, evaluated identically. Bold values across this row indicate per-victim peaks.

Epoch	Ckpt	MATH500	AIME24	AIME25	JEE Math (s/p)	LCB pass@1
0	base (Qwen2.5-7B-Instruct)	71.0	8.9	2.2	32.2 / 35.9	15.8
1	step-00625	63.1	7.8	2.2	— / —	—
2	step-01250	67.6	10.0	6.7	— / —	—
3	step-01875	72.9	14.4	8.9	35.7 / 39.3	17.6
4	step-02500	75.8	14.4	13.3	35.2 / 39.5	19.0
5	step-03125	75.9	13.3	13.3	35.0 / 39.9	15.8

Training recipe

• Base model: Qwen/Qwen2.5-7B-Instruct
• Teacher traces: Chia-Mu-Lab/openthoughts_distill_victim_data_10k_clean_swag
• Recipe: s1 paper exact full fine-tune (FSDP full-shard, no LoRA)
• Block size: 32768 tokens · effective batch 16 (mb=1, ga=4, 4×H200)
• Optimizer: AdamW, lr=1e-5 cosine, warmup_ratio=0.05, bf16
• Epochs: 5, save_strategy=epoch

Files

ot-q3_14b-clean/
  README.md
  metrics.csv          ← machine-readable per-epoch metric table
  epoch_1/             ← full HF checkpoint dir (config.json, model-*.safetensors,
  epoch_2/                tokenizer*, etc.)
  epoch_3/
  epoch_4/
  epoch_5/

Caveats / known issues

• All epochs here are from the canonical s1-distill 3-exp sweep (2026-05-20), evaluated with the unified math500+AIME+JEE+LCB scorer pipeline.
• JEE Math here refers to the subject="math" subset (≈236 of 515 questions) scored per the official dair-iitd compute_metrics.py. The strict number is the headline accuracy; partial gives MCQ(multiple) partial credit.
• These models are research artifacts for the steel-reasoning-trace project (reasoning-trace extraction attack study); do not use for production.