LoRA adapter for Qwen/Qwen3.6-27B, trained on 40% difficult-advice / 60% TULU3 replay with loss on assistant tokens only.

Controlled ablation of …-tulu-lora-40-60: same mixture file (md5 88f39a3d01e59ba9d592b26c1705c57f, 1,982 rows, 1,493,359 tokens), same seed, same hyperparameters, same rendered strings. The loss mask is the only difference.

Training data: qwen3.6-27b-sft-mixture-40-60_assistant_loss_only, published with per-row assistant spans so the mask reproduces without our code.

Supervision starts immediately after the <|im_start|>assistant\n header — which the model is given at inference and never has to produce — and ends after the closing <|im_end|>, which it must produce in order to stop. Everything else is -100.

<|im_start|>   MASKED
assistant      MASKED
\n             MASKED
<think>        LOSS     <- supervision starts at the first generated token

TRL's assistant_only_loss flag does not work on this model. It requires {% generation %} markers the chat template lacks, and it re-renders from messages, discarding 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; TRL receives finished labels.

Zero rows carry an empty <think></think> — that pattern is Qwen3.6's explicit do-not-deliberate marker and trains a model to stop reasoning.

bf16 LoRA (not QLoRA — bitsandbytes does not reliably cover this model's hybrid linear-attention/SSM layers: 48 of 64 layers are Gated DeltaNet, and none of their projections receive an adapter, so quantization error there would be uncorrectable). 1×H100 80GB, 90 min.

Final train loss 0.948, token accuracy 0.749. Loss is not comparable to the full-token arm by construction — a different set of tokens is scored.

Not yet evaluated. For reference, the full-token sweep at the same total budget:

The 40% row is this adapter's direct counterpart.

Sibling assistant-loss-only arms: 10-90 · 20-80 · 40-60

Use AutoModelForImageTextToText, not AutoModelForCausalLM — this is a vision-language checkpoint. Merging drops the base model's 15 mtp.* tensors, so speculative decoding needs them grafted back.