Results (all on held-out, freshly generated instances)

Per-query ensemble completeness (same instances, every possible query — the latents serve ALL queries from one fixed computation, including threads never named in any text):

• Final value of chain 1: : 1.000
• Final value of chain 2: : 0.993
• Final value of chain 3: : 1.000
• Chain number with the largest final value: : 0.047
• Chain number with the smallest final value: : 0.147
• Last digit of the sum of the three final values: : 0.813

Training data is an infinite procedural stream (every instance seen at most once); "train accuracy" evaluates instances bit-exactly replayed from the training seed, "test" a disjoint seed — the match shows the organism runs the algorithm rather than memorizing instances.

Probe findings (ridge linear probes on dot residual streams)

raw workspace, not an answer cache: finals v1/v2/v3 decode at ~1.0 across the block but the SUM is at chance (0.06-0.16) at every dot - yet behavioral sum accuracy is 0.81, so the aggregate is assembled downstream in the masked suffix from raw latent operands. Training history: suffix-side aggregation over a latent operand needed ~3x the stage budget to train (v2 stalled at the first latent operand; a dot-side sum variant never trained at all).

Worked example

See examples.md for a full operating transcript (the model sees the scenario, emits only dots in <think>, then the query is revealed and it answers from the dot activations) and the surface-CoT version of the same instance that the curriculum started from.

Files

• LoRA adapter (PEFT, r=32, attn+MLP target modules) + tokenizer + lt_cfg.json (full config)
• training_code/ — complete training/eval/probe code snapshot (see its README)
• plots/ — load-bearing eval, probe grid + logit lens, training curve

Intended use

Activation-oracle / interpretability research target: the reasoning trace exists ONLY in the dot activations (logit lens is near-dark), with deterministic ground truth from the task generator. Wired into the cds-jb/AVBench recog eval as suite latent_threads (token-exact rows; row_metadata.selected=False rows query threads with no textual trace anywhere in the transcript). Part of the latent-threads collection together with its four siblings.

Trained 2026-06-12 (wandb group lt3g, MATS10-CS-JB/cot-oracle). Built on the bottleneck-mask methodology of the pointer-chase filler organism (cds-jb/qwen3-8b-pointer-chase-filler-cot).