LoRA adapter for Qwen/Qwen3.6-27B, trained on 10% difficult-advice / 90% TULU3 replay (~1M tokens) with loss on assistant tokens only, for 1 epoch.

The difficult-advice data comes from synthdoc_v2, a stage-for-stage replication of the Teaching Claude Why difficult-advice pipeline — six separate stages rather than the collapsed generate-then-revise loop of the original synthdoc.

Training data: qwen3.6-27b-synthdocv2-mixture-10_90.

Drawn from a pool holding 25 examples for each of the 8 constitutional principles. Each difficult-advice record carries a real <think> reasoning trace; TULU3 replay rows carry no think block at all, and no row anywhere carries an empty <think></think>.

Everything outside an assistant turn is -100; 78.3% of tokens carry loss. A supervised span starts immediately after the <|im_start|>assistant\n header and ends after the closing <|im_end|>, which the model must emit in order to stop.

TRL's own assistant_only_loss cannot do this on Qwen3.6 — its chat template has no {% generation %} markers, and TRL re-renders from messages, which would discard the think-block convention baked into the pre-rendered mixture. Spans are derived from the rendered text via the fast tokenizer's offset mapping instead.

bf16 LoRA (not QLoRA — bitsandbytes does not reliably cover this model's hybrid linear-attention/SSM layers). 1×H100 80GB, 54 min.

Final train loss 0.901, token accuracy 0.811.

Not yet evaluated on ODCV-Bench or agentic-misalignment.

Sibling arms: 10/90 · 15/85 · 20/80

Use AutoModelForImageTextToText, not AutoModelForCausalLM — this is a vision-language checkpoint.