{
  "schema": "cadence-attention-heads-founder-control/1",
  "declared_date": "2026-10-22",
  "scope": "Founder numerical-policy control on the unchanged raw dual-binding component: actual Brain.compose dt=1.0 with existing bounded halving damping=3, complete original-field qualification, no new learning rule or initializer. Capacity first; no learned-selection, head-advantage, semantic or continuing-life claim from acquisition.",
  "source": {
    "baseline": "607d4e80",
    "freeze": "Commit this protocol and every executed source before the one capacity stage; retain all earlier failed sources and receipts. Later relevance requires this exact positive capacity source manifest."
  },
  "development_seeds": [
    0,
    1
  ],
  "confirmation_seeds": [
    200,
    201,
    202,
    203,
    204,
    205,
    206,
    207,
    208,
    209,
    210,
    211,
    212,
    213,
    214,
    215,
    216,
    217,
    218,
    219
  ],
  "confirmation_required": 18,
  "confirmation_status": "unopened; requires all development gates and a separately declared total budget",
  "task": {
    "inputs": 19,
    "names": [
      "A",
      "B"
    ],
    "shared_values": [
      0,
      1,
      2,
      3
    ],
    "markers": [
      "IS",
      "NEUTRAL",
      "AMBIGUOUS_CUE",
      "ASK_AB",
      "ASK_BA"
    ],
    "coordinates": {
      "markers": [
        0,
        1,
        2,
        3,
        4
      ],
      "names": [
        5,
        6
      ],
      "values": [
        7,
        8,
        9,
        10
      ],
      "exposed_A_values": [
        11,
        12,
        13,
        14
      ],
      "exposed_B_values": [
        15,
        16,
        17,
        18
      ]
    },
    "facts": "IS + name + value; two actual fact events, both orders balanced, then one neutral event",
    "query": "AMBIGUOUS_CUE + ASK_AB or ASK_BA; no values or intermediate address",
    "output_slots": [
      4,
      4
    ],
    "answers": "(value_A,value_B) for ASK_AB; (value_B,value_A) for ASK_BA",
    "held_out_value_pairs": [
      [
        0,
        1
      ],
      [
        1,
        0
      ],
      [
        2,
        3
      ],
      [
        3,
        2
      ]
    ],
    "training_pairs": "All remaining 12 ordered pairs; all equal pairs and every value under each name remain in training",
    "training_cases": 48,
    "held_out_cases": 16,
    "neutral_events": 1,
    "rounds": 512,
    "batch": 16,
    "schedule": "SeedSequence([622022, founder]); each round samples 16 training cases with replacement; every arm uses identical draws",
    "evaluation": "Enumerate all training and held-out cases once; exact pair, each slot, unequal pairs and each query reported; refused/unrun answers count wrong",
    "swapped_panel": "Exchange values attached to A/B with query, fact order and token multiset fixed; held-out swaps stay held out; equal-value swaps are shams"
  },
  "brain": {
    "modules": [
      32
    ],
    "episodic": false,
    "lateral": 0.0,
    "recall_next": false,
    "recall_address": 0.0,
    "recall_null": 0.0,
    "recall_recency": 0.0,
    "recall_amplitude": 3.0,
    "recall_temperature": 0.25,
    "teaching": "normalized",
    "learning_qualified": true,
    "learning_damping": 3,
    "learning_free_steps": 2048,
    "learning_nudged_steps": 2048,
    "learning_beta": 0.1,
    "learning_temperature": 0.2,
    "learning_tolerance": 0.003,
    "learning_momentum": 0.9,
    "empty_anchor": true
  },
  "dt": 1.0,
  "teaching": "Capture complete pre-query drive and immutable bank before greedy act; wrong-pair rows only, both final labels supplied; no head role, slot index, address, intermediate value or relevance mask taught. Corresponding rows of captured drive and bank remain together through free/nudged phases.",
  "primary_boundary": "Clear working trace and warm neural state after prefix before query in training/evaluation, keep only witnessed memory bank. Live query/read/association/motor coordinates all qualify jointly. Intact-trace comparison is later and separate.",
  "heads": "Shared bank; disjoint live query/read ports; same complete query initial drive; query-cortex projections zero initially, ordinary reciprocal read-cortex seams independently drawn. Total coupling3 shared equally over independent heads; no supplied head-to-name assignment.",
  "arms": [
    {
      "name": "exposed",
      "kind": "plain_Brain",
      "modules": [
        32
      ],
      "input": "both unselected named values in dormant control coordinates"
    },
    {
      "name": "supplied_reads",
      "heads": 2,
      "modules": [
        32
      ],
      "memory_energy": "Held-address agreement with empty_anchor=true: alpha1.5 per head,total3; zero-content read agreement persists on empty streams; no external read source.",
      "input": "Actual witnessed fact_A/fact_B held as oracle target in respective read agreement; no external3*fact drive, no ask-dependentselector, no address teacher."
    },
    {
      "name": "H1",
      "heads": 1,
      "modules": [
        32
      ],
      "coupling_per_head": 3.0
    },
    {
      "name": "H2",
      "heads": 2,
      "modules": [
        32
      ],
      "coupling_per_head": 1.5
    },
    {
      "name": "H1-width",
      "heads": 1,
      "modules": [
        64
      ],
      "coupling_per_head": 3.0
    },
    {
      "name": "H1-port-copies",
      "selectors": 1,
      "port_copies": 2,
      "modules": [
        32
      ],
      "total_coupling": 3.0,
      "representation": "Duplicate keys/values/cues scaled by 1/sqrt(2); same total query/read coordinates as H2, one selector, unchanged match/read norm"
    },
    {
      "name": "H1-work",
      "heads": 1,
      "modules": [
        32
      ],
      "coupling_per_head": 3.0,
      "maximum_sweeps_per_phase": 4096,
      "comparison": "Measured full neuron+transport work matched to H2; unmatched work gives no efficiency/head-specific claim"
    },
    {
      "name": "frozen-shuffled",
      "kind": "acquired-copy selection-interface interventions and frozen newborn controls"
    },
    {
      "name": "uniform_random",
      "seed": 622023,
      "expected_exact_pair": 0.0625
    }
  ],
  "candidate_order": [
    "agreement"
  ],
  "maximum_candidate_rounds": 1,
  "candidate_selection": "After all four capacity cells pass, one separate agreement-only relevance round across all five H1/H2/resource arms and founders0/1; all ten cells remain. No linear fallback, initializer, parameter grid or confirmation.",
  "gates": {
    "exposed_acquisition_floor": 0.95,
    "exposed_panel": "All 48 training cases: exact pair, each slot, unequal-value pairs and each query; both development founders. The 16 held-out cases are reported separately.",
    "content_use_floor": 0.95,
    "content_use": "Held-address agreement oracle on H2 anatomy, both actual namedfacts, totalcoupling3 and same restoring law; finalpair/slot/unequal/query95 training48 bothfounders, held16reported. No learned-head assignment claim.",
    "held_out_exact_pair_floor": 0.9,
    "swapped_unequal_exact_pair_floor": 0.9,
    "relevant_record_erasure_slot_loss_minimum": 0.2,
    "donor_pair_follow_minimum": 0.8,
    "sham_or_record_reordering_answer_change_maximum": 0.05,
    "retention_loss_maximum": 0.05,
    "retention_rounds": 256,
    "retention": "Continue balanced training on the same 12 pairs, then read the held-out panel; retain .95 on the training panel. No competing-task or P7 claim.",
    "causal_controls": [
      "queried-record erasures",
      "irrelevant-neutral-record change",
      "both-values donor transplant",
      "one-head erasures",
      "duplicated heads",
      "head permutation with downstream transport custody explicit",
      "frozen/shuffled acquired selection interface"
    ],
    "promotion": "All development gates must pass on both founders 0 and 1 before confirmation; retain failures, refusals and caps. No library or default promotion from this experiment."
  },
  "representation_phase": {
    "start": "Only after primary acquisition/content/retrieval/causal gates pass",
    "variable": "Free settled association activation, not context-free held LearnedCode output",
    "probe": "Two diagnostic ridge readouts to one-hot requested values, fit on trainingpairs only; centered features, ridge0.01; freeze before held-out and swapped panels; never installed as answer path or teaching rule",
    "pair_transfer_floor": 0.9,
    "gain_over_raw_query_and_frozen_newborn_minimum": 0.2,
    "loss_after_relevant_memory_erasure_minimum": 0.2,
    "claim": "Task-context accessibility and held-out readout transfer, not general meaning",
    "controls": [
      "current raw query only",
      "fixed random observation code separately labeled if run",
      "frozen newborn association",
      "record erasure",
      "same ambiguous cue with swapped witnessed facts"
    ],
    "later_semantic_gate": "Separately freeze learned-code recipe/source/budget and compare raw, fixed-random, learned, no-top-down; require held-out alias/group members without raw overlap, misleading raw overlap, member-specific facts and early facts after competition. Geometric shifts alone do not pass."
  },
  "budget": {
    "serial_local_processes": 1,
    "training_seconds_per_arm": 120,
    "development_total_host_seconds": 360,
    "additional_stage": true,
    "scope": "additional_founder_capacity_stage",
    "charged": "All construction, initial/final evaluation, training, actual attempted phases/transports and refusals in this new stage. Earlier charges are reported cumulatively, not subtracted from this deadline. Boundaries are checked between rounds/panels; retain bounded final-panel overrun.",
    "stop": "One frozen founder numerical-policy control only. Capped, refused and unrun cells cannot establish admission; no cap, dt or force grid and no fresh confirmation seeds."
  },
  "preservation": "Retain previous failed admission/coupled-reader receipts and existing gates. Reward custody, private imagination, complete checkpoint continuation, continuing context/recovery and learned semantic codes remain separate requirements before integration.",
  "capacity_control": "held-address-agreement",
  "numerical_control": {
    "base_protocol_file": "heads-fixed-read-protocol.json",
    "base_dt": 0.1,
    "dt": 1.0,
    "damping": 3,
    "default_guard": "Fresh plain Brain.compose without an explicit neuron-model or dt override must report dt1.0. learning_neuron_model helper default0.5 and bare NeuronModel default0.2 are not the composed founder.",
    "reason": "Saved failed-boundary Jacobians are locally unstable for Euler dt1.0, while dt0.5 is locally stable at those sampled points; use the existing bounded halving policy with exact immutable memory preserved. Some earlier accepted states have a positive continuous mode or a kink; no uniform contraction or global convergence claim. Full original-field qualification remains mandatory.",
    "initializer": "The separately frozen finite memory-start initializer failed both saved boundaries; it is not implemented or used.",
    "equations": "Same continuous field, fixed witnessed bank and learned parameters. Numerical dt-only clones preserve the actual graph/memory, and all free/nudged states are tested against the complete original equations."
  },
  "prior": {
    "receipt_file": "empty-boundary-development.json",
    "receipt_kind": "cadence-attention-empty-boundary-admission/1",
    "receipt_digest": "d3afc60bb45e3468cbe0d878c530d37ae283d262d2348df5d192374430b25f14",
    "receipt_cumulative_host_seconds": 404.31181683630786,
    "failure_diagnostic_host_seconds": 94.9148939170409,
    "initializer_diagnostic_host_seconds": 1.1108592920936644,
    "algebra_audit_host_seconds": 1.0,
    "previous_host_seconds": 501.3375700454424,
    "custody": "Require the retained failed receipt via --prior; verify digest, arithmetic and original archived source bytes with the repository verifier. Historical sources need not equal the new numerical-policy sources."
  },
  "relevance_stage": {
    "additional_stage": true,
    "development_total_host_seconds": 600,
    "training_seconds_per_arm": 120,
    "capacity_receipt_file": "founder-capacity-development.json",
    "required": "All four current-source capacity cells pass unchanged .95 trained floors and complete512 rounds without refusals before any of the ten agreement-only relevance cells. Preserve cumulative prior and capacity charges; this new600-second stage has an independent deadline. No confirmation or default promotion."
  },
  "capacity_stage": {
    "order": [
      "exposed:0",
      "exposed:1",
      "supplied_reads:0",
      "supplied_reads:1"
    ],
    "guard": "Both exposed founders must pass before the supplied-read pair runs. An exposed failure stops admission before oracle work; missing/unrun oracle cells can never produce a positive capacity result. All four cells must pass before any relevance work.",
    "gate": "Unchanged .95 on complete48 trained cases for exact pair, each slot, unequal pairs and each query,512 rounds and no action/lesson refusal. Final intact train/held_out/all panels must be fully answered without refusal or budget truncation; the16 held-out cases have no accuracy floor and are reported separately. Optional erased-panel accuracy is not an acquisition gate."
  }
}
