out-of-core rework: embed_cluster_acts streams resids to disk memmap + streaming multi-GPU k-means (--kmeans-sample); build_data reads unified emb.f32 + GPU-batched Newton/IRLS probes

build_data: hard-negative pairing in Qwen3 residual space (--hard-neg-k; subtle probe directions)

build_data: hold out probe-fit docs from SFT targets (no direction<->text contamination)

maxact-fast: hardening fixes (6 bugs) + Dr.GRPO vllm-lens rl.py + sbatch; Probe = universal direction

build_data: real validated LR probes (held-out AUC + activation-margin filter)

maxact-fast: clean repo — DESIGN + pretraining pipeline (embed/cluster/probe-cache/build-data/pretrain), Dr. GRPO RL to follow