bi_traversal_thought_graph.py

#!/usr/bin/env python3
"""
Bi-Traversal Thought Graph System
==================================

A production-ready implementation for tracing AI "thoughts" from axiom anchors
to contextual conclusions using bidirectional graph traversal.

Based on: "Bi-Traversal Pathways — from Axiom Anchor → End Contextual Aligned Piece"

Author: Claude (based on provided blueprint)
Version: 1.0.0
Date: 2025-10-20
License: MIT
"""

import uuid
import json
import numpy as np
from typing import Dict, List, Tuple, Optional, Any, Set
from dataclasses import dataclass, field, asdict
from datetime import datetime
from collections import defaultdict, deque
import networkx as nx
from pathlib import Path as FilePath

try:
    from sentence_transformers import SentenceTransformer
    SENTENCE_TRANSFORMERS_AVAILABLE = True
except ImportError:
    SENTENCE_TRANSFORMERS_AVAILABLE = False
    print("Warning: sentence-transformers not installed. Using fallback embeddings.")
    print("Install with: pip install sentence-transformers")


# ============================================================================
# Data Structures
# ============================================================================

@dataclass
class NodeMetadata:
    """Metadata for a graph node"""
    confidence: float = 1.0
    provenance: str = ""
    timestamp: str = field(default_factory=lambda: datetime.now().isoformat())
    logical_form: Optional[str] = None
    token_span: Optional[Tuple[int, int]] = None
    model_id: Optional[str] = None
    custom: Dict[str, Any] = field(default_factory=dict)

    def to_dict(self) -> Dict:
        return asdict(self)


@dataclass
class Node:
    """Represents a cognitive node in the thought graph"""
    id: str = field(default_factory=lambda: str(uuid.uuid4()))
    type: str = "concept"  # axiom | concept | op | evidence | subaxiom | context | surface_text
    text: str = ""
    embedding: Optional[np.ndarray] = None
    metadata: NodeMetadata = field(default_factory=NodeMetadata)
    is_anchor: bool = False

    def to_dict(self) -> Dict:
        return {
            'id': self.id,
            'type': self.type,
            'text': self.text,
            'embedding': self.embedding.tolist() if self.embedding is not None else None,
            'metadata': self.metadata.to_dict(),
            'is_anchor': self.is_anchor
        }


@dataclass
class Edge:
    """Represents an inference edge between nodes"""
    from_id: str
    to_id: str
    type: str = "inference"  # inference | transform | reference | implication | counterexample
    weight: float = 0.9
    rule_id: Optional[str] = None
    metadata: Dict[str, Any] = field(default_factory=dict)

    def to_dict(self) -> Dict:
        return asdict(self)


@dataclass
class Path:
    """Represents a path through the thought graph"""
    nodes: List[Node]
    edges: List[Edge]
    score: float = 0.0
    metadata: Dict[str, Any] = field(default_factory=dict)

    def node_ids(self) -> List[str]:
        return [n.id for n in self.nodes]

    def to_dict(self) -> Dict:
        return {
            'nodes': [n.to_dict() for n in self.nodes],
            'edges': [e.to_dict() for e in self.edges],
            'score': self.score,
            'metadata': self.metadata
        }


# ============================================================================
# Embedding Engine
# ============================================================================

class EmbeddingEngine:
    """Handles semantic embeddings for nodes"""

    def __init__(self, model_name: str = 'all-MiniLM-L6-v2'):
        self.model_name = model_name
        if SENTENCE_TRANSFORMERS_AVAILABLE:
            self.model = SentenceTransformer(model_name)
            self.dimension = self.model.get_sentence_embedding_dimension()
        else:
            # Fallback: simple hash-based embeddings
            self.model = None
            self.dimension = 384
            print(f"Using fallback embeddings (dimension={self.dimension})")

    def encode(self, text: str, normalize: bool = True) -> np.ndarray:
        """Generate embedding for text"""
        if self.model is not None:
            return self.model.encode(text, normalize_embeddings=normalize)
        else:
            # Simple fallback: hash-based pseudo-embedding
            return self._fallback_embedding(text, normalize)

    def encode_batch(self, texts: List[str], normalize: bool = True) -> np.ndarray:
        """Generate embeddings for multiple texts"""
        if self.model is not None:
            return self.model.encode(texts, normalize_embeddings=normalize)
        else:
            return np.array([self._fallback_embedding(t, normalize) for t in texts])

    def _fallback_embedding(self, text: str, normalize: bool = True) -> np.ndarray:
        """Simple hash-based embedding for when sentence-transformers unavailable"""
        # Use hash to seed random embedding
        np.random.seed(hash(text) % (2**32))
        emb = np.random.randn(self.dimension)
        if normalize:
            emb = emb / np.linalg.norm(emb)
        return emb

    @staticmethod
    def cosine_similarity(a: np.ndarray, b: np.ndarray) -> float:
        """Compute cosine similarity between two embeddings"""
        return float(np.dot(a, b) / (np.linalg.norm(a) * np.linalg.norm(b)))


# ============================================================================
# Thought Graph
# ============================================================================

class ThoughtGraph:
    """Main graph structure for thought representation"""

    def __init__(self, embedding_engine: Optional[EmbeddingEngine] = None):
        self.graph = nx.DiGraph()
        self.embedding_engine = embedding_engine or EmbeddingEngine()
        self.nodes_by_id: Dict[str, Node] = {}
        self.edges_by_id: Dict[Tuple[str, str], Edge] = {}

    def add_node(self, node: Node, auto_embed: bool = True) -> str:
        """Add a node to the graph"""
        if auto_embed and node.embedding is None:
            node.embedding = self.embedding_engine.encode(node.text)

        self.nodes_by_id[node.id] = node
        self.graph.add_node(
            node.id,
            node_obj=node,
            type=node.type,
            text=node.text,
            embedding=node.embedding,
            is_anchor=node.is_anchor
        )
        return node.id

    def add_edge(self, edge: Edge) -> None:
        """Add an edge to the graph"""
        self.edges_by_id[(edge.from_id, edge.to_id)] = edge
        self.graph.add_edge(
            edge.from_id,
            edge.to_id,
            edge_obj=edge,
            type=edge.type,
            weight=edge.weight,
            rule_id=edge.rule_id
        )

    def get_node(self, node_id: str) -> Optional[Node]:
        """Retrieve a node by ID"""
        return self.nodes_by_id.get(node_id)

    def get_edge(self, from_id: str, to_id: str) -> Optional[Edge]:
        """Retrieve an edge by endpoints"""
        return self.edges_by_id.get((from_id, to_id))

    def get_anchor_nodes(self) -> List[Node]:
        """Get all anchor nodes"""
        return [n for n in self.nodes_by_id.values() if n.is_anchor]

    def get_neighbors(self, node_id: str, direction: str = 'forward') -> List[str]:
        """Get neighbor node IDs"""
        if direction == 'forward':
            return list(self.graph.successors(node_id))
        else:  # backward
            return list(self.graph.predecessors(node_id))

    def find_similar_nodes(self, embedding: np.ndarray, threshold: float = 0.82, limit: int = 10) -> List[Tuple[Node, float]]:
        """Find nodes with similar embeddings"""
        results = []
        for node in self.nodes_by_id.values():
            if node.embedding is not None:
                sim = self.embedding_engine.cosine_similarity(embedding, node.embedding)
                if sim >= threshold:
                    results.append((node, sim))

        # Sort by similarity descending
        results.sort(key=lambda x: x[1], reverse=True)
        return results[:limit]

    def to_dict(self) -> Dict:
        """Export graph to dictionary"""
        return {
            'nodes': [n.to_dict() for n in self.nodes_by_id.values()],
            'edges': [e.to_dict() for e in self.edges_by_id.values()],
            'metadata': {
                'node_count': len(self.nodes_by_id),
                'edge_count': len(self.edges_by_id),
                'timestamp': datetime.now().isoformat()
            }
        }

    def save_json(self, filepath: FilePath) -> None:
        """Save graph to JSON file"""
        with open(str(filepath), 'w') as f:
            json.dump(self.to_dict(), f, indent=2)


# ============================================================================
# Bi-Traversal Engine
# ============================================================================

class BiTraversalEngine:
    """Implements bidirectional graph traversal for thought tracing"""

    def __init__(self, graph: ThoughtGraph):
        self.graph = graph

    def frontier_expand(
        self,
        start_nodes: List[str],
        direction: str = 'forward',
        max_steps: int = 6,
        top_k: int = 5
    ) -> List[Tuple[Node, Path]]:
        """
        Expand frontier from starting nodes

        Args:
            start_nodes: List of node IDs to start from
            direction: 'forward' or 'backward'
            max_steps: Maximum path length
            top_k: Keep top K paths

        Returns:
            List of (end_node, path) tuples
        """
        # Priority queue: (negative_score, path)
        frontier = []

        # Initialize with start nodes
        for node_id in start_nodes:
            node = self.graph.get_node(node_id)
            if node:
                path = Path(nodes=[node], edges=[], score=1.0)
                frontier.append((-1.0, path))

        completed_paths = []
        visited_nodes: Set[str] = set()

        for step in range(max_steps):
            new_frontier = []

            for neg_score, path in frontier:
                current_node = path.nodes[-1]

                # Mark as visited
                visited_nodes.add(current_node.id)

                # Get neighbors
                neighbor_ids = self.graph.get_neighbors(current_node.id, direction)

                if not neighbor_ids:
                    # Dead end - add to completed paths
                    completed_paths.append((current_node, path))
                    continue

                # Expand to neighbors
                for neighbor_id in neighbor_ids:
                    if neighbor_id in visited_nodes:
                        continue

                    neighbor = self.graph.get_node(neighbor_id)
                    if not neighbor:
                        continue

                    # Get edge
                    if direction == 'forward':
                        edge = self.graph.get_edge(current_node.id, neighbor_id)
                    else:
                        edge = self.graph.get_edge(neighbor_id, current_node.id)

                    if not edge:
                        continue

                    # Create new path
                    new_path = Path(
                        nodes=path.nodes + [neighbor],
                        edges=path.edges + [edge],
                        score=path.score * edge.weight * neighbor.metadata.confidence
                    )

                    new_frontier.append((-new_path.score, new_path))

            # Keep top K paths
            new_frontier.sort(key=lambda x: x[0])
            frontier = new_frontier[:top_k]

            if not frontier:
                break

        # Add remaining frontier paths to completed
        for neg_score, path in frontier:
            completed_paths.append((path.nodes[-1], path))

        # Sort by score and return top K
        completed_paths.sort(key=lambda x: x[1].score, reverse=True)
        return completed_paths[:top_k]

    def bi_traverse(
        self,
        anchor_id: str,
        end_signature: str,
        k: int = 5,
        tau: float = 0.82,
        max_steps: int = 6
    ) -> List[Path]:
        """
        Perform bidirectional traversal from anchor to end signature

        Args:
            anchor_id: Starting anchor node ID
            end_signature: Target text signature
            k: Number of paths to return
            tau: Similarity threshold for matching
            max_steps: Maximum steps in each direction

        Returns:
            List of complete paths sorted by score
        """
        # Generate embedding for end signature
        end_embedding = self.graph.embedding_engine.encode(end_signature)

        # Find candidate end nodes
        end_candidates = self.graph.find_similar_nodes(end_embedding, threshold=tau, limit=k*2)
        end_node_ids = [node.id for node, sim in end_candidates]

        if not end_node_ids:
            print(f"      Warning: No end nodes found matching signature with threshold {tau}")
            print(f"      Trying with all nodes as fallback...")
            # Fallback: try all surface_text nodes
            end_node_ids = [nid for nid, n in self.graph.nodes_by_id.items() if n.type == 'surface_text']
            if not end_node_ids:
                print(f"      Error: No surface_text nodes found")
                return []

        # Forward expansion from anchor
        print(f"[*] Forward expansion from anchor {anchor_id}...")
        forward_frontier = self.frontier_expand(
            start_nodes=[anchor_id],
            direction='forward',
            max_steps=max_steps,
            top_k=k
        )

        # Backward expansion from end nodes
        print(f"[*] Backward expansion from {len(end_node_ids)} end candidates...")
        backward_frontier = self.frontier_expand(
            start_nodes=end_node_ids,
            direction='backward',
            max_steps=max_steps,
            top_k=k
        )

        # Meet in the middle
        print(f"[*] Finding intersection points...")
        matches = self._find_matches(forward_frontier, backward_frontier, tau)

        # Score and sort
        scored_paths = []
        for path in matches:
            score = self._compute_path_score(path, end_embedding)
            path.score = score
            scored_paths.append(path)

        scored_paths.sort(key=lambda p: p.score, reverse=True)
        return scored_paths[:k]

    def _find_matches(
        self,
        forward_frontier: List[Tuple[Node, Path]],
        backward_frontier: List[Tuple[Node, Path]],
        tau: float
    ) -> List[Path]:
        """Find matching nodes between forward and backward frontiers"""
        matches = []

        for fnode, fpath in forward_frontier:
            for bnode, bpath in backward_frontier:
                # Check if nodes match (by ID or similarity)
                if fnode.id == bnode.id:
                    # Exact match
                    merged_path = self._merge_paths(fpath, bpath, reverse_backward=True)
                    matches.append(merged_path)
                elif fnode.embedding is not None and bnode.embedding is not None:
                    sim = self.graph.embedding_engine.cosine_similarity(
                        fnode.embedding,
                        bnode.embedding
                    )
                    if sim >= tau:
                        # Similar match
                        merged_path = self._merge_paths(fpath, bpath, reverse_backward=True)
                        merged_path.metadata['match_similarity'] = sim
                        matches.append(merged_path)

        return matches

    def _merge_paths(self, forward_path: Path, backward_path: Path, reverse_backward: bool = True) -> Path:
        """Merge forward and backward paths at intersection point"""
        if reverse_backward:
            # Reverse backward path (it was traced backwards)
            b_nodes = list(reversed(backward_path.nodes[:-1]))  # Exclude last node (intersection)
            b_edges = list(reversed(backward_path.edges))
        else:
            b_nodes = backward_path.nodes[:-1]
            b_edges = backward_path.edges

        return Path(
            nodes=forward_path.nodes + b_nodes,
            edges=forward_path.edges + b_edges,
            score=forward_path.score * backward_path.score,
            metadata={
                'forward_score': forward_path.score,
                'backward_score': backward_path.score,
                'intersection_node': forward_path.nodes[-1].id
            }
        )

    def _compute_path_score(self, path: Path, end_embedding: np.ndarray,
                           alpha: float = 0.5, beta: float = 0.35,
                           gamma: float = 0.15, delta: float = 0.4) -> float:
        """
        Compute comprehensive path score

        Score = α·log_score + β·semantic_coherence + γ·provenance - δ·penalty
        """
        # Log score (cumulative edge weights and node confidences)
        log_score = 0.0
        for edge in path.edges:
            log_score += np.log(max(edge.weight, 0.01))
        for node in path.nodes:
            log_score += np.log(max(node.metadata.confidence, 0.01))

        # Normalize log score
        normalized_log_score = 1.0 / (1.0 + np.exp(-log_score / len(path.nodes)))

        # Semantic coherence (path end vs target)
        if path.nodes[-1].embedding is not None:
            semantic_coherence = self.graph.embedding_engine.cosine_similarity(
                path.nodes[-1].embedding,
                end_embedding
            )
        else:
            semantic_coherence = 0.5

        # Provenance bonus (nodes with explicit provenance)
        provenance_scores = []
        for node in path.nodes:
            if node.metadata.provenance:
                provenance_scores.append(1.0)
            else:
                provenance_scores.append(0.5)
        provenance_bonus = np.mean(provenance_scores) if provenance_scores else 0.5

        # Logical penalty (assumptions without evidence)
        assumptions = sum(1 for n in path.nodes if n.type in ['concept', 'op'] and not n.metadata.provenance)
        logical_penalty = assumptions * 0.1

        # Combine
        final_score = (
            alpha * normalized_log_score +
            beta * semantic_coherence +
            gamma * provenance_bonus -
            delta * logical_penalty
        )

        return max(0.0, min(1.0, final_score))


# ============================================================================
# English Explanation Generator
# ============================================================================

class ExplanationGenerator:
    """Converts graph paths to readable English explanations"""

    def __init__(self):
        self.node_type_verbs = {
            'axiom': 'starts from',
            'concept': 'considers',
            'op': 'applies',
            'evidence': 'observes',
            'context': 'contextualizes',
            'surface_text': 'concludes'
        }

    def explain_path(self, path: Path, include_provenance: bool = True) -> Dict[str, Any]:
        """
        Generate English explanation for a path

        Returns:
            Dict with 'concise' and 'detailed' explanations
        """
        sentences = []

        # Process each step
        for i, node in enumerate(path.nodes):
            sentence = self._node_to_sentence(node, i, len(path.nodes))
            if include_provenance and node.metadata.provenance:
                sentence += f" (provenance: {node.metadata.provenance}, confidence: {node.metadata.confidence:.2f})"
            sentences.append(sentence)

        # Concise version
        concise_sentences = self._compress_sentences(sentences)
        concise = ' '.join(concise_sentences)

        # Detailed version with numbering
        detailed_lines = []
        for i, sent in enumerate(sentences):
            detailed_lines.append(f"{i+1}. {sent}")
        detailed = '\n'.join(detailed_lines)

        # Add metadata summary
        metadata_summary = self._format_metadata(path)

        return {
            'concise': concise,
            'detailed': detailed,
            'metadata': metadata_summary,
            'score': path.score
        }

    def _node_to_sentence(self, node: Node, index: int, total: int) -> str:
        """Convert a node to a sentence fragment"""
        verb = self.node_type_verbs.get(node.type, 'processes')

        if index == 0:
            return f"Starting from the {node.type} '{node.text}'"
        elif index == total - 1:
            return f"therefore {node.text}"
        else:
            return f"{verb} '{node.text}'"

    def _compress_sentences(self, sentences: List[str]) -> List[str]:
        """Compress sentences for concise version"""
        if len(sentences) <= 3:
            return sentences

        # Keep first, last, and most important middle
        middle_text = sentences[len(sentences)//2]
        if "'" in middle_text:
            middle_excerpt = middle_text.split("'")[1]
        else:
            middle_excerpt = middle_text[:50]

        compressed = [
            sentences[0],
            f"through {len(sentences) - 2} intermediate steps including '{middle_excerpt}'",
            sentences[-1]
        ]
        return compressed

    def _format_metadata(self, path: Path) -> str:
        """Format path metadata as readable text"""
        parts = [
            f"Path length: {len(path.nodes)} nodes",
            f"Confidence score: {path.score:.3f}",
        ]

        if 'intersection_node' in path.metadata:
            parts.append(f"Intersection at: {path.metadata['intersection_node'][:8]}...")

        if 'match_similarity' in path.metadata:
            parts.append(f"Match similarity: {path.metadata['match_similarity']:.3f}")

        return ' | '.join(parts)


# ============================================================================
# Example Builder (for demonstrations)
# ============================================================================

def build_example_graph() -> ThoughtGraph:
    """
    Build the example graph from the blueprint:
    Observer_loop_stabilize → phase_lock_144000 → model_update
    """
    graph = ThoughtGraph()

    # Create nodes
    anchor = Node(
        type='axiom',
        text='0D Anchor: Observer_loop_stabilize',
        is_anchor=True,
        metadata=NodeMetadata(
            confidence=1.0,
            provenance='system_initialization',
            model_id='claude-sonnet-4-5'
        )
    )

    perception = Node(
        type='op',
        text='perception_event: sample_stream',
        metadata=NodeMetadata(
            confidence=0.95,
            provenance='telemetry_subsystem',
            logical_form='sample(telemetry_stream, t)'
        )
    )

    phase_sig = Node(
        type='evidence',
        text='phase_signature_144000',
        metadata=NodeMetadata(
            confidence=0.85,
            provenance='FFT_analyzer_output',
            token_span=(1000, 1050)
        )
    )

    state_update = Node(
        type='op',
        text='state_update_rule_X: trigger_on_phase_lock',
        metadata=NodeMetadata(
            confidence=0.90,
            provenance='control_system_v2',
            logical_form='update(internal_state) IF phase_lock(144000)'
        )
    )

    conclusion = Node(
        type='surface_text',
        text='AI updates internal model when phase lock is detected at 144000 cycles',
        metadata=NodeMetadata(
            confidence=0.88,
            provenance='reasoning_engine',
            model_id='claude-sonnet-4-5'
        )
    )

    # Add nodes to graph
    for node in [anchor, perception, phase_sig, state_update, conclusion]:
        graph.add_node(node)

    # Create edges
    edges = [
        Edge(anchor.id, perception.id, type='inference', weight=0.95, rule_id='observer_perception_link'),
        Edge(perception.id, phase_sig.id, type='transform', weight=0.90, rule_id='FFT_detection'),
        Edge(phase_sig.id, state_update.id, type='implication', weight=0.92, rule_id='phase_lock_trigger'),
        Edge(state_update.id, conclusion.id, type='inference', weight=0.88, rule_id='explanation_synthesis')
    ]

    for edge in edges:
        graph.add_edge(edge)

    return graph


# ============================================================================
# CLI / Demo Interface
# ============================================================================

def run_demonstration():
    """Run the complete demonstration from the blueprint"""
    print("="*80)
    print("  Bi-Traversal Thought Graph - Demonstration")
    print("="*80)

    # Build example graph
    print("\n[1/4] Building example thought graph...")
    graph = build_example_graph()
    print(f"      Graph created: {len(graph.nodes_by_id)} nodes, {len(graph.edges_by_id)} edges")

    # Get anchor
    anchors = graph.get_anchor_nodes()
    if not anchors:
        print("ERROR: No anchor nodes found")
        return
    anchor = anchors[0]

    # Perform bi-traversal
    print(f"\n[2/4] Performing bi-traversal from anchor '{anchor.text[:40]}...'")
    engine = BiTraversalEngine(graph)

    end_signature = "AI updates internal model when phase lock is detected at 144000 cycles"
    paths = engine.bi_traverse(
        anchor_id=anchor.id,
        end_signature=end_signature,
        k=3,
        tau=0.60,  # Lower threshold for fallback embeddings
        max_steps=6
    )

    print(f"      Found {len(paths)} complete paths")

    # Generate explanations
    print(f"\n[3/4] Generating English explanations...")
    explainer = ExplanationGenerator()

    for i, path in enumerate(paths):
        explanation = explainer.explain_path(path, include_provenance=True)

        print(f"\n" + "-"*80)
        print(f"  Path {i+1} (Score: {explanation['score']:.3f})")
        print("-"*80)

        print(f"\nConcise Explanation:")
        print(f"  {explanation['concise']}")

        print(f"\nDetailed Explanation:")
        for line in explanation['detailed'].split('\n'):
            print(f"  {line}")

        print(f"\nMetadata:")
        print(f"  {explanation['metadata']}")

    # Export
    print(f"\n[4/4] Exporting results...")
    output_path = FilePath("thought_graph_output.json")
    graph.save_json(output_path)
    print(f"      Graph saved to: {output_path}")

    # Export best path
    if paths:
        best_explanation = explainer.explain_path(paths[0], include_provenance=True)
        explanation_path = FilePath("best_explanation.json")
        with open(str(explanation_path), 'w') as f:
            json.dump(best_explanation, f, indent=2)
        print(f"      Best explanation saved to: {explanation_path}")

    print("\n" + "="*80)
    print("  Demonstration Complete")
    print("="*80)


if __name__ == "__main__":
    run_demonstration()

thought_graph_output.json

{
  "nodes": [
    {
      "id": "6243f0ef-5fc4-4900-8ee1-146d8625bf77",
      "type": "axiom",
      "text": "0D Anchor: Observer_loop_stabilize",
      "embedding": [
        -0.023467466753783823,
        -0.014599497159624684,
        -0.04053848546490672,
        -0.10906144150924167,
        -0.04767798762157954,
        -0.05002547056080734,
        0.05946285692760377,
        -0.10165157733346583,
        0.021513292047978676,
        -0.11970891758367973,
        0.13723301111371136,
        -0.10457866904396787,
        -0.06673052922053332,
        0.08751161133776843,
        -0.01757211074548442,
        -0.0033113013526364743,
        0.04343458971640608,
        -0.018124306886783157,
        -0.03849919717220384,
        -0.14589087959745678,
        0.10414996851011127,
        -0.04071587255865248,
        0.002351168456980892,
        -0.0017656258339032512,
        -0.09821853717989835,
        0.02209490496471462,
        0.00017934765594931201,
        -0.025218369908620707,
        -0.12308018513392223,
        -0.04976343763627332,
        0.0018542661654270488,
        0.07552797394753288,
        -0.13753333681228427,
        0.05195493387616278,
        0.00464897245120283,
        0.09491596635121143,
        -0.01646209551031087,
        -0.12065820317413396,
        -0.040452842021662175,
        -0.021592219879018187,
        0.023223343159154603,
        0.0008233450832208623,
        -0.06319573968256584,
        0.002934516212561422,
        -0.0054262987418381954,
        -0.07155333981030192,
        -0.015191456052493646,
        -0.011062473486569701,
        0.0027402957626784137,
        0.0024013926114016437,
        -0.028720917117840885,
        0.009019671718908683,
        0.012762652568241455,
        0.017020382246752997,
        0.059739962188673425,
        -0.01066985913404868,
        -0.0013897876312953892,
        -0.008369070562172582,
        -0.05975153342478306,
        0.032815150895443464,
        -0.005852808482264905,
        -0.06479302770900444,
        -0.026486604110598354,
        -0.010713839169995718,
        -0.06125795524948102,
        0.039496473608993894,
        0.10937287085494647,
        0.03071354098997769,
        -0.06387354040640199,
        0.03562830412723235,
        0.02673963932237353,
        -0.004528163181568574,
        -0.029207440906540313,
        0.03228830416714888,
        0.028531509284157246,
        -0.008766840884376943,
        -0.019906148531570775,
        -0.07471270129224214,
        0.011833287383822517,
        0.014943994331510224,
        0.066141423781385,
        0.0414027139527738,
        -0.06292486121697108,
        0.011145064105862697,
        -0.06885819120271373,
        0.02507183904725603,
        0.0018577333681472964,
        0.08963196057437803,
        0.025433500749738505,
        -0.010669724459730518,
        0.11419301395119755,
        0.007185780321602748,
        -0.06379724075549142,
        0.06671928379114957,
        0.06949083995558421,
        -0.002812947398640205,
        0.003395943106330467,
        0.019441109611516193,
        -0.00637392285111268,
        -0.05453659246968743,
        0.05172297615931056,
        -0.054174038509051184,
        -0.00036280882641673084,
        -0.07639894176658837,
        0.015741681329333,
        0.03645622475385367,
        -0.04206869055992417,
        -0.026513030644290902,
        -0.006320597058081801,
        0.014846057205044783,
        -0.07975724528885976,
        0.01870233255457172,
        0.05425937525561539,
        -0.018605025959141187,
        -0.040220000178559354,
        0.04300621550655164,
        -0.0005762921346117074,
        0.06731657939385824,
        -0.038550541375050776,
        0.06676232311618262,
        -0.027658403385738217,
        -0.04473652237063253,
        0.044084595478084375,
        -0.004675656256599518,
        -0.0422115214451742,
        0.039177716694531156,
        -0.003483619448296199,
        0.03875057886253887,
        -0.09210880768632773,
        -0.013202842573849568,
        0.03052693071780755,
        0.010426415934912258,
        -0.03478294300405905,
        -0.011853608440362477,
        0.10489846104154661,
        -0.01343632454749719,
        -0.010754778847170543,
        -0.0946014264704632,
        0.09253806326529322,
        0.009566917063774948,
        -0.02885447006150916,
        0.0690692484769335,
        0.049106201115714745,
        0.03524558692856217,
        -0.05924773298929312,
        -0.03110614959175983,
        0.003401113941764275,
        -0.11390017256118201,
        -0.05155920599039024,
        0.034965570725564755,
        0.05663133333827101,
        0.03482219819350386,
        0.02231131519200625,
        -0.04333238331446509,
        -0.05746456448659299,
        0.03349817712013067,
        -0.017670339814676642,
        0.02887986674862267,
        -0.08042833806311558,
        -0.017307039810724466,
        -0.030348579344508252,
        -0.02481606385437269,
        0.059598022644116115,
        0.016066846816283988,
        -0.09722209759355871,
        -0.02434445943304833,
        -0.031500782489181256,
        -0.020783833455063433,
        0.04146612516862908,
        -0.1017461333762452,
        0.005541951649672894,
        -0.037539005695356,
        -0.0650531639536349,
        -0.10548455112911051,
        -0.03824439879506211,
        0.03234921135507963,
        -0.04941469611590305,
        -0.009513100311474202,
        0.09028738408896883,
        0.04334162450873908,
        0.01771712411103772,
        -0.06242324129282804,
        0.03412891090216149,
        0.05237627677932872,
        -0.00015660115770045,
        -0.054821102697872645,
        -0.033227785007739166,
        0.01351068414194305,
        0.037692680445258926,
        0.04829970617668366,
        -0.018053195120055203,
        -0.023646708318723383,
        0.030318996469163445,
        0.009127626251255868,
        -0.007311378277539145,
        0.08150767808293188,
        -0.00015818175937078386,
        -0.0013609951226809139,
        -0.038245054584857996,
        -0.012158572877610607,
        0.027494752670774242,
        -0.08071093848231745,
        0.0458905793772618,
        -0.05709502228251169,
        0.08552256232350403,
        -0.016429760111006386,
        -0.003001101968489037,
        -0.02767577677679023,
        -0.05338328421074393,
        -0.05048138203941256,
        -0.045893079746212165,
        -0.06463256623000631,
        0.03195401516472003,
        -0.01229319909740903,
        0.0017107673838848386,
        -0.07936888265277603,
        0.006633650920101835,
        0.055690538598693516,
        0.011922900469528975,
        0.13707270872458532,
        0.030338293540979836,
        -0.08226593383521676,
        0.014897845636161283,
        0.030392591131443682,
        -0.05327232119242935,
        0.04114855498224026,
        -0.01576522899139752,
        -0.05907184312571806,
        -0.03324602020665973,
        0.012042510674614705,
        -0.03718758150034032,
        -0.08000565762473155,
        0.04185036595885972,
        0.00024070688417203331,
        -0.006271131797773084,
        -0.018378919629972124,
        -0.009480020833606334,
        -0.02466645280347229,
        0.0031080868964202587,
        0.012500427969876531,
        -0.03206309140790048,
        -0.04698246825309129,
        0.024793373233090853,
        -0.06396002759523411,
        -0.08455285584132288,
        -0.033430307129838105,
        -0.029102546012392688,
        0.06944694548930738,
        -0.029523571821466525,
        -0.063214844025699,
        -0.05537662597740951,
        0.0704094676253914,
        0.0170201693913237,
        -0.018691301099166998,
        -0.028446948170366415,
        -0.03436757742483569,
        0.09713841890705303,
        -0.046144289375520314,
        0.06310183463736938,
        -0.048571813906575154,
        0.03702680014994222,
        0.04700953182633008,
        -0.017992131134377174,
        0.07030447616329019,
        0.08423431501638658,
        0.05439711968373177,
        0.008139351108317384,
        0.00984758792146654,
        -0.03224503936102978,
        -0.044187331099672655,
        0.11644044787724311,
        -0.004704996985493438,
        0.004293889294593455,
        0.06453410309631141,
        0.011416367516558902,
        -0.05749008903306439,
        0.01323034126734774,
        0.035862412071321355,
        0.032692100418803674,
        0.008194027864212682,
        -0.003551985994770029,
        0.03274329947410193,
        -0.0008890996789847934,
        0.007553219546187134,
        -0.013494057295956673,
        -0.04897391584149689,
        0.05545323978073976,
        -0.00523371652584114,
        -0.024133794605948534,
        0.02858446063476828,
        0.04589132708139076,
        -0.015114313565906312,
        -0.0885230287020931,
        0.012243046284169543,
        -0.025285722742173115,
        0.04079586161290278,
        0.05502014696267198,
        -0.03391331543490928,
        -0.07041348477193503,
        -0.00975754360562698,
        0.037074123611818784,
        0.0695135937568361,
        0.056603676501206956,
        0.028050121372537696,
        0.02323600760374045,
        -0.019518952789389917,
        0.025484819422953862,
        -0.06652218721040971,
        -0.08507816421780609,
        0.024325948912764718,
        -0.009467122039735737,
        0.04831276144891938,
        0.013551626964123677,
        0.005214631144406427,
        -0.019120949238334625,
        0.016289379460375005,
        0.11031420139776665,
        -0.00021418772873234534,
        -0.04243306124731699,
        -0.0524037568513859,
        -0.05127970653693164,
        0.009009489655827842,
        0.006398725661215325,
        -0.04775932898824558,
        -0.0806091202796243,
        0.05780000025411568,
        0.08832249361020413,
        -0.11043846778119765,
        -0.029672067224203597,
        0.0006754658587062531,
        -0.028408497735934946,
        0.08079248035788052,
        -0.03327745078273466,
        0.013959384068463272,
        0.11118761967491561,
        -0.04218373282220805,
        0.05035300238020383,
        -0.05492069543843918,
        0.014898534600479367,
        0.013159275572316274,
        -0.025561308141285793,
        -0.04449401947358386,
        0.052318445791754514,
        -0.014980096929215298,
        -0.014531390780964486,
        0.03589303093287462,
        0.08875388023488859,
        -0.009509156321256934,
        0.1039985751985369,
        -0.06894287728612443,
        -0.00019211099618440935,
        -0.024141854439002262,
        0.019581001923298678,
        0.08423975803827463,
        -0.02746636932275455,
        -0.03635942820447487,
        -0.03073748604512529,
        0.047130862360530316,
        -0.08486860462789933,
        -0.022998793875574713,
        0.024657634675212495,
        0.0175908149499861,
        0.13813697914234582,
        -0.02170207269304502,
        -0.05879717948563457,
        0.06697740700841373,
        0.04443841413418544,
        0.08812997626625646,
        -0.033903007514825736,
        -0.006018604735021891,
        -0.0033106353227485836,
        0.0009301631803589754,
        -0.05073759703007293,
        0.0028966780433150045,
        0.06949275042322944,
        0.06437022454873055,
        -0.02463430566597067,
        0.06682196353501484,
        0.04547964162671852,
        0.055413456973819546,
        -0.00846108890268665,
        0.02527271925971485,
        -0.04333909194921056,
        0.07660716037619231
      ],
      "metadata": {
        "confidence": 1.0,
        "provenance": "system_initialization",
        "timestamp": "2025-10-20T03:03:42.442097",
        "logical_form": null,
        "token_span": null,
        "model_id": "claude-sonnet-4-5",
        "custom": {}
      },
      "is_anchor": true
    },
    {
      "id": "b7fe5a4c-3826-4926-9c8f-6ee16aa352f1",
      "type": "op",
      "text": "perception_event: sample_stream",
      "embedding": [
        0.034147982754480985,
        -0.007162431285976829,
        -0.02988300165623102,
        -0.03791215843144269,
        -0.06993277992979914,
        0.02821237936215313,
        -0.01442175148029419,
        -0.013968131732726638,
        0.01718468170581747,
        -0.021349590588658384,
        -0.02948552967392239,
        -0.03538251551649593,
        -0.03930817250142916,
        -0.06652610075853098,
        -0.07857786087611758,
        0.0627835286127621,
        0.0025291471989493204,
        -0.03409430915644847,
        0.033689085569818755,
        0.01944046342450329,
        0.010164890973472684,
        -0.040429895531032406,
        -0.027504508311313475,
        -0.017528951816244295,
        0.0747241258481132,
        -0.022318789946986214,
        -0.05810594522142377,
        -0.0399246868851914,
        0.06593323093399203,
        0.05563742439738363,
        -0.027217632052289248,
        0.0315646303250003,
        -0.04148878446134887,
        0.07011961698882133,
        -0.008704337622947153,
        -0.03328371699772853,
        -0.04402401291366972,
        0.02245254796030957,
        0.08699968236042818,
        -0.024100953392040353,
        -0.07746485346043241,
        -0.033578913650144766,
        -0.020437541205411386,
        -0.019803970293408784,
        0.01906412093865649,
        -0.04232798553724495,
        -0.10368951466731592,
        0.06033969403325768,
        0.02356013003732911,
        -0.0005392688449206675,
        -0.009978940277881408,
        0.02294741221047607,
        0.10840784545592953,
        -0.03772521572704846,
        0.04421217659386817,
        0.021196721645032633,
        0.07489631084910925,
        -0.013496198407271596,
        0.03140076255172302,
        -0.02189175749880231,
        -0.0017878458915092208,
        -0.047367592946044215,
        -0.050334245228632175,
        0.003400370338703189,
        -0.06472570051798969,
        -0.06542363108668613,
        -0.0469542286686953,
        -0.05436115814118983,
        -0.015463389268347371,
        0.03962540777817674,
        -0.09835837992695858,
        0.03172127902620952,
        -0.01978725288660417,
        -0.03883720455651777,
        0.09411204857147124,
        0.03906348400838904,
        -0.030459378129634347,
        -0.05798833666287039,
        -0.015409624045425268,
        -0.03511483818105962,
        -0.03349111856852764,
        0.0061706390768988895,
        0.05950550861970101,
        0.0372282634797712,
        -0.031769994006189864,
        -0.006765319420061291,
        0.06403610918179335,
        0.01517188050882689,
        0.06668773585382214,
        0.05173758591393396,
        -0.0365150027865752,
        -0.008223483784087722,
        0.03868343838263557,
        0.018361101334318952,
        -0.0033443062432176327,
        0.03597752883523871,
        -0.07992924542285347,
        -0.02737215022348809,
        -0.025974556353072633,
        -0.06035858652254435,
        -0.013305141883821729,
        -0.10487037008746376,
        -0.09680626786944328,
        -0.02700269128680042,
        -0.008753920637458938,
        0.04001548503029659,
        0.08255057292382273,
        0.021340751154187137,
        0.015069051885881634,
        -0.05051745731599974,
        -0.010859048129887822,
        -0.007474209129192105,
        0.12394269845896835,
        -0.015496258187623492,
        -0.014879221920714643,
        0.036751928599890876,
        -0.06302618708606357,
        0.030948961550656053,
        -0.06742795711164626,
        0.15592228537622765,
        0.07074159353318997,
        -0.011057179212505475,
        0.01716444441616603,
        0.0427519887796263,
        -0.016346151711311357,
        0.023220647210169337,
        -0.1269187682540371,
        0.010885120099581832,
        -0.06346035343523043,
        0.02955024972029039,
        0.028395447895817747,
        -0.03330662638977513,
        0.007267619991986594,
        -0.03165327829010838,
        -0.013899688345844291,
        -0.021444867320014216,
        0.022185697212420537,
        0.006412405991103943,
        -0.014828917202062737,
        0.06278110020106058,
        -0.010993049656260344,
        -0.04628756567514494,
        -0.02478882876292077,
        -0.06379647723343816,
        0.04671529376193871,
        -0.02810508402614528,
        0.045486397761325285,
        0.005600937571036561,
        -0.007079435961121854,
        0.03823236185346304,
        0.004479218438649979,
        -0.025935353625938164,
        -0.005945384813615097,
        -0.054208303092733556,
        -0.019529162981846992,
        -0.061492726339890814,
        0.021312671932788112,
        -0.012261208465039329,
        0.07509783756541039,
        0.04847845730403576,
        0.05199415168082681,
        0.0680800716254034,
        0.07920164921670386,
        0.046539246580053956,
        0.013428338228154065,
        -0.05344955463823496,
        -0.021848012513215644,
        0.04087956041856243,
        -0.0042294757517221115,
        -0.017269241055565376,
        0.06406885846495144,
        -0.04926814846523737,
        -0.006618993215597161,
        -0.018282626075789937,
        -0.06896793214288556,
        0.006185005421953222,
        -0.03766352197789802,
        -0.00520467721800186,
        0.06002600957245588,
        -0.08399142714901071,
        -0.06574256764577739,
        0.09344160379891356,
        -0.03139650306540798,
        -0.03718016396425912,
        0.07070365786677825,
        -0.07264369254604212,
        0.013942574137745849,
        -0.015596497662619556,
        0.077428149745543,
        0.0027497562352631977,
        -0.12192269913529487,
        0.07072106180316781,
        -0.04427383424307023,
        -0.018914742333540898,
        -0.014009432807484758,
        -0.016524844097740174,
        -0.007088772477622479,
        0.151614778195266,
        0.0396271142623743,
        -0.12258745248051973,
        0.0975575875960479,
        0.01543021513932059,
        -0.006883275065164997,
        0.03010099405492495,
        -0.049053958824435216,
        0.017691748195391495,
        -0.040986120845388975,
        -0.09146911243758657,
        -0.12582226224640874,
        0.1052646365400338,
        -0.05521493723117616,
        -0.00018600274184913163,
        0.0008325362412781712,
        0.052345826131209174,
        -0.03483854940813793,
        0.0033149787784918054,
        0.0028892672515826914,
        0.01822440177026305,
        0.06341752912026907,
        -0.020419993229720147,
        -0.0519513545219371,
        0.037747884238478115,
        -0.07082781597283454,
        -0.07863550077725427,
        -0.043578721246971715,
        0.007190236622118739,
        0.03906406955396244,
        -0.04771453645268234,
        -0.0760461327340263,
        -0.017194577911205655,
        -0.010783117740527846,
        -0.020728474533700962,
        -0.027561272720774268,
        -0.15759196252993865,
        0.007295725255314128,
        -0.05175878383190917,
        0.0940594286533772,
        -0.037496582101378656,
        0.02803112205097458,
        -0.00509760845818683,
        0.01785349235923327,
        0.053680324466651615,
        -0.02687512308102842,
        0.006622888762346537,
        -0.022303423917624132,
        0.021630085588500127,
        0.009660128195749887,
        0.10485403427334782,
        0.060497952626951575,
        0.0016180576486365967,
        0.0013571703368132862,
        -0.020697012958689514,
        -0.0017071036093425527,
        -0.029934131533944804,
        0.052366383947530196,
        0.03002567194045568,
        0.007876970186987055,
        -0.07129237373553614,
        -0.09367263237166429,
        -0.03852270900896401,
        0.08048011999931863,
        0.07842505026790474,
        0.06816518987593179,
        0.07008884384899022,
        -0.12665842089084606,
        -0.0896615673667431,
        0.04095525561717767,
        0.035081397899925065,
        -0.011464436766692445,
        0.017354195671660768,
        -0.031147435226735767,
        -0.011438224442756408,
        0.06609359738368308,
        0.08139727767167206,
        0.017679759392357573,
        -0.007461179201351363,
        0.04394907479367614,
        -0.07315336658305077,
        0.02311336501504154,
        -0.07559663324187843,
        0.032958678220852286,
        0.09133530875588335,
        0.05797230500309755,
        0.049694476232773704,
        -0.02602287002382351,
        -0.016616125890006855,
        -0.0028763512424463183,
        -0.044382287533876484,
        -0.02212876117706477,
        -0.016675198256430077,
        -0.02957314466544656,
        -0.010575315813780085,
        -0.019857325175000735,
        -0.0917045025469571,
        -0.005470904117492301,
        -0.06920439260256399,
        0.09152265435290823,
        0.04532102068148163,
        -0.028289455741084372,
        0.022209396252458092,
        -0.018193758158301144,
        0.036449872031798516,
        0.12007896293446106,
        -0.023644618296369775,
        -0.04766518618366488,
        -0.03133335382916835,
        0.05729810333942529,
        -0.03916033707321709,
        -0.036688193985933364,
        -0.07526358277961952,
        0.05853312034832641,
        -0.005864549295067179,
        0.02699134828633762,
        0.02132098532242145,
        -0.03624939288835819,
        0.035533394640004275,
        0.01318606851571909,
        -0.0949382846131316,
        -0.08520060648199504,
        -0.03844731712019388,
        0.0006863475072705642,
        -0.11478340943256578,
        -0.05854641393537082,
        0.013358687668932397,
        -0.04289057412316495,
        -0.04468299796409731,
        0.06483389213516808,
        0.08160628762425437,
        -0.05139342016655931,
        -0.05652648603682023,
        -0.007214299234942649,
        0.020733908370267035,
        0.010033891247953544,
        -0.0044547809569396545,
        0.04922286556068909,
        0.020509922722229353,
        -0.020404548172269017,
        0.049081716759901146,
        0.10013942941561102,
        -0.0733491704937565,
        -0.06733026127406379,
        -0.01778858594511104,
        -0.053561321653275276,
        -0.06807668091273761,
        0.05454106672797688,
        -0.04095906190003433,
        0.031646973180307375,
        0.0005197622810286547,
        -0.00778204376152131,
        -0.030472317646060086,
        -0.06887419395611039,
        -0.11979228378975434,
        -0.06460183367989905,
        -0.040953261344175144,
        -0.007295119415592881,
        0.009579137669560533,
        0.007858708647028211,
        -0.01759470701681118,
        -0.05278777655179856,
        -0.011504239533163447,
        -0.10408679819840072,
        -0.04031356570608089,
        -0.06511263288259708,
        0.018151840892579953,
        -0.05499272407387338,
        0.009563310195653213,
        -0.078046892586059,
        0.013181054929295563,
        -0.07413176071282375,
        -0.01706913940210535,
        0.041280507242551924,
        -0.017627605213328314,
        0.050941968468389436,
        0.01097232124810105,
        -0.005151590351594485,
        -0.03380517133747711,
        -0.0023080949908329917,
        -0.012568266207130523,
        0.009101615173920333,
        -0.0778332506217458,
        0.013062548432434767,
        -0.017342318445466742,
        -0.01288880699444965,
        0.07491335143903576
      ],
      "metadata": {
        "confidence": 0.95,
        "provenance": "telemetry_subsystem",
        "timestamp": "2025-10-20T03:03:42.442137",
        "logical_form": "sample(telemetry_stream, t)",
        "token_span": null,
        "model_id": null,
        "custom": {}
      },
      "is_anchor": false
    },
    {
      "id": "9a4ff851-2ef3-4b22-8454-1f6c1e9d0e00",
      "type": "evidence",
      "text": "phase_signature_144000",
      "embedding": [
        -0.012967622353573044,
        0.01091056559414843,
        -0.007000824905304681,
        0.037849177374878924,
        0.024386596217791993,
        0.04299734945594247,
        -0.009354911410626634,
        -0.05642988001415192,
        -0.004979338861212211,
        -0.016324878190028584,
        0.06838902834985042,
        -0.04588251427741704,
        -0.06756661159489494,
        0.001978406896986248,
        0.03507556468969499,
        -0.0392233511608612,
        -0.059857077077787674,
        0.045687996329877516,
        0.11103877625934709,
        -0.02771319528662319,
        0.004982591240222326,
        -0.05782156056660488,
        -0.05448043800377611,
        0.03031657834882986,
        -0.00724957205417757,
        -0.04626767666171554,
        0.04518178944527538,
        0.0018882704144869882,
        -0.015368120135697457,
        0.03514538433077634,
        -0.025104359563273546,
        -0.007930970031812926,
        -0.03124425570694865,
        -0.004764622504349888,
        -0.03346306390506051,
        -0.06562352261579422,
        -0.004669309232990687,
        -0.04493221201680404,
        0.08709094141141045,
        0.04893396272711379,
        -0.11844766154405213,
        -0.03760604888714692,
        -0.016485382713018623,
        0.05659569993674703,
        -0.007871726651384425,
        0.006352303940253659,
        0.009378431718718517,
        0.06331783132620457,
        0.07786673482033978,
        0.0004043177192283644,
        0.07750818106704885,
        0.072642765461256,
        0.015716808919885403,
        -0.07157732648333619,
        0.029967016597304278,
        0.01654065391089422,
        -0.005205171677492773,
        -0.011618903930026457,
        0.08169996409687759,
        0.08532448316691385,
        0.022927412582741127,
        -0.04386247124724328,
        0.0484656999503153,
        -0.012503632755695698,
        -0.01589691677451013,
        0.015307584316970404,
        -0.08908805693289074,
        0.007936816320557464,
        0.05087943948370305,
        -0.08074984051964405,
        -0.043961330724021405,
        0.05446768277185618,
        -0.10076301845331888,
        -0.01180490538905659,
        -0.032823292873034655,
        -0.016715342634104527,
        0.05003433343222953,
        -0.013315785793174951,
        0.019647513486450614,
        0.01586883292368341,
        0.043626972110540896,
        -0.03731554393965477,
        0.07861073446484093,
        0.09854219703602762,
        0.07717903748379061,
        -0.07456154120595121,
        -0.08983150428674425,
        -0.07288430178208188,
        -0.007672197697945101,
        -0.01628757323016821,
        -0.02571208800038583,
        0.03474272524570778,
        -0.015178999266201657,
        -0.019017574951759545,
        0.06432728504628227,
        -0.012648801320394403,
        -0.038129809399757936,
        -0.016431190551361437,
        0.1500105031751433,
        0.0012045562963607832,
        -0.012242551503763797,
        -0.0430952705330049,
        -0.04499463567150201,
        -0.019464158768948057,
        -0.003204289116427417,
        0.04883333545232274,
        -0.029522251494814786,
        0.0031802939128549148,
        -0.02951830990692166,
        -0.03774499800015915,
        -0.07717130533297684,
        -0.08601771877469087,
        0.07447824209562132,
        0.002370920560226796,
        0.06788895848306395,
        0.044145309571125,
        0.08115539476556316,
        -0.008918247593760633,
        -0.12413082428335208,
        0.02307922411610165,
        0.04677431773340376,
        -0.018459742436688367,
        -0.07498705843632526,
        0.022557050133129097,
        -0.010681701313556568,
        -0.033500920205719986,
        0.043299142191133205,
        0.06863816191687298,
        0.022365037700199854,
        0.003754727175101133,
        0.042733292932199664,
        -0.06947899936417985,
        0.00023536829437865263,
        -0.05293750783376669,
        0.0165594104843886,
        -0.05770375347145606,
        0.06297399105600437,
        -0.032361522689836504,
        0.07448530233354955,
        -0.021188703610730497,
        -0.00234374916094881,
        0.06290977834598059,
        -0.10942517007413194,
        0.03894567753395505,
        -0.08550898210081785,
        0.024829006292603303,
        -0.0656552441299251,
        -0.04668157726090829,
        -0.060318373951970275,
        0.06085653633165506,
        -0.015003783739307495,
        -0.06477893539446351,
        -0.039106281369583676,
        0.02860345960928236,
        -0.03572993162038697,
        -0.018651469173083666,
        -0.028014343130533706,
        -0.014365538033215589,
        -0.009058684241343755,
        0.0669325416571437,
        -0.0526976503964259,
        0.08740168121431141,
        -0.03402579624696848,
        -0.002991101797561388,
        0.04385060194386852,
        -0.05544419839166813,
        0.06286782289793562,
        0.002197605770811441,
        -0.04947859075648837,
        -0.0021199506236570112,
        -0.058261598193396796,
        -0.038260989696896945,
        -0.0483298682876377,
        0.12428563741606731,
        -0.014727831651289361,
        -0.052022714652881594,
        0.15644680213441695,
        0.03922864106455532,
        -0.02335845293650705,
        -0.03361626602319008,
        -0.0011499333682500501,
        -0.0018994404487160765,
        -0.04809008179125845,
        0.02033750258345929,
        -0.10607680289444497,
        -0.022581063605925024,
        -0.040008536940921526,
        -0.047263294516064105,
        0.05015167279304853,
        0.0552527837943679,
        -0.043147896651562545,
        -0.020845842738936048,
        0.0643461435608884,
        -0.04803884321313072,
        -0.006120924790419991,
        -0.014189108995448882,
        -0.026498801647467242,
        0.08702630554901955,
        0.05000367522399881,
        -0.08280793490996509,
        0.10849192869903891,
        0.09318627339369412,
        0.06596432439614004,
        -0.06321155178465451,
        -0.023784007246601253,
        0.034824940302632425,
        0.05426729253641051,
        0.04512992073774798,
        0.007965881301213347,
        -0.03994139378640537,
        -0.0030058444060299264,
        0.057187122858318104,
        -0.006775985791940908,
        -0.005635004629074619,
        -0.02726895211211364,
        -0.022448690407544738,
        0.0456733087093819,
        -0.07387323137178381,
        -0.1298726259279311,
        -0.01671817273785452,
        0.044096332324878924,
        0.0769270735834976,
        -0.05544192730649267,
        -0.03923754407420264,
        0.006888772877305173,
        0.003414088983270342,
        0.05073234805817988,
        0.023810579885248137,
        0.02608978813234972,
        0.07472768261942123,
        -0.043225761242387854,
        -0.03480508291757661,
        0.07230109323103139,
        -0.0213459055507761,
        -0.16075210138823165,
        -0.0012367255261378914,
        -0.06637763771290844,
        -0.02005691665805148,
        0.07690551354445586,
        -0.03798618048810956,
        -0.005170808334154515,
        0.048107869788561644,
        0.0400453240214582,
        0.06329616246609356,
        0.0652431957915006,
        0.044928992252531885,
        0.07664618290972118,
        0.01496841131739428,
        0.03758762454331315,
        0.011406303463459413,
        0.07616948542666944,
        0.014633196837317525,
        -0.09170416421916837,
        -0.0198593366258219,
        0.0016255001035830792,
        0.004698510360698294,
        -0.16143740265412865,
        -0.06546939927395833,
        -0.020401249820220308,
        0.030926632137542223,
        0.03807747606225072,
        -0.04081918718115205,
        0.03990114929929367,
        0.040373494638316194,
        0.00033479162165404657,
        0.07032669580479378,
        -0.0005699287395054078,
        0.05259191032136765,
        0.07348251795254734,
        0.036050470382972336,
        0.055841727887168784,
        -0.05358246972007377,
        0.059580662349310184,
        0.003020476133682588,
        -0.0689738621230649,
        0.036542693052372496,
        -0.00030412232772911483,
        -0.05382399849715931,
        0.020185407368338945,
        -0.020277804354757827,
        -0.02033385295693747,
        -0.022147004487861843,
        -0.0020346133912133055,
        -0.026273890446395783,
        -0.014563896950263333,
        -0.09874643540722643,
        0.001318615113833018,
        -0.023746024800350533,
        -0.06002455813738519,
        0.005415612646398997,
        0.005367062843084356,
        0.13237393169676703,
        0.023308395566030333,
        0.022726926355268687,
        -0.06538071051747546,
        0.06928721732951247,
        0.041349522378453525,
        -0.07849561680252523,
        0.0033969469144674263,
        -0.04718487677685312,
        -0.010739366291476255,
        0.01990216658681278,
        -0.0518673725201348,
        0.04635846555544206,
        0.0689632833772998,
        -0.04847709523952142,
        0.058900610881491414,
        0.025220596998829146,
        -0.04861303972943651,
        0.021670793027687374,
        -0.0540129949905305,
        0.07484986982820027,
        -0.03014267362780353,
        0.015485829253334023,
        -0.026143658627607706,
        -0.008571968272567072,
        0.03571361965878625,
        0.011163036231101013,
        0.06306771293961645,
        0.013421525244262614,
        0.04444159592624086,
        -0.0820412321221119,
        -0.0069584639572677146,
        0.08731296442266875,
        0.00887695518266039,
        -0.07454498020789753,
        -0.005570360925660519,
        -0.023187719061318363,
        0.07123715229405712,
        0.056051303821088194,
        0.001705227482820114,
        0.03348241121315201,
        0.0511399740653833,
        0.013421888910427848,
        -0.01201073325335439,
        -0.0916828181995901,
        -0.01461915912586281,
        -0.07116702493068401,
        -0.07773563508767864,
        0.04671832005592244,
        0.005288336824015635,
        0.012315595991860943,
        -0.044499058103456716,
        -0.009391517259443224,
        0.017760553935145112,
        -0.04983812672966449,
        -0.015051439015170434,
        -0.013480632675947986,
        -0.049503635900267955,
        -0.010421729542194228,
        0.020749803831376954,
        -0.05736427161068285,
        -0.007723844726547399,
        -0.020921710381344785,
        0.01238970042680505,
        -0.05590370634183482,
        -0.044731864246843715,
        0.021294552706998295,
        0.04246151180541373,
        0.06180660694319762,
        -0.01454836750910405,
        -0.06464933184394026,
        -0.05062597498813857,
        -0.05824741785610044,
        0.028369170313079995,
        0.050477267827894294,
        0.048506670728467836,
        0.01850617450516413,
        -0.05844953857751393,
        0.008511059963933382,
        -0.051210289605530784,
        -0.03279725356931907,
        -0.012929782258171072,
        0.026883183448743325,
        -0.10979074299697988,
        0.015245945489727823,
        -0.013678842653879562,
        0.09780816910342366,
        0.06758347118651496,
        -0.041733050407060976,
        -0.04886553546158213,
        0.015733216904688232,
        -0.017523720146377472,
        -0.02077871983289286
      ],
      "metadata": {
        "confidence": 0.85,
        "provenance": "FFT_analyzer_output",
        "timestamp": "2025-10-20T03:03:42.442146",
        "logical_form": null,
        "token_span": [
          1000,
          1050
        ],
        "model_id": null,
        "custom": {}
      },
      "is_anchor": false
    },
    {
      "id": "99f50fc8-941f-4a94-b2fc-bb9d74d424c2",
      "type": "op",
      "text": "state_update_rule_X: trigger_on_phase_lock",
      "embedding": [
        0.004592711274599245,
        0.003470470929233585,
        -0.03730687115353603,
        -0.03440065342611229,
        0.06179263391088655,
        -0.06500960848736997,
        -0.05603726114766621,
        -0.019967147832040618,
        0.010662767634295597,
        0.056349204864135936,
        0.004460709794674605,
        0.009542654777035309,
        -0.0033024741943943415,
        -0.018082651016300876,
        0.011716036238365464,
        -0.015578712585655054,
        0.08180745331376116,
        0.016174911419860695,
        0.007798656181523412,
        0.06707162412684123,
        -0.02095217912574201,
        0.027557638987066226,
        -0.055393618722801546,
        0.04890084921970781,
        0.09978611855943241,
        0.02351935234661824,
        0.025872723575926893,
        0.05814770487554458,
        0.10672839030641482,
        -0.10442733644772881,
        0.0328880639177179,
        0.09877750658956543,
        0.0290340305774238,
        -0.03956004821262799,
        0.005843579095993293,
        -0.029394334038972566,
        -0.007076589073042456,
        0.08138576651152576,
        -0.044670595917008386,
        0.06742255949741453,
        -0.04915226562101377,
        -0.01730573058824882,
        -0.0450971720158866,
        0.03390704821573625,
        -0.008786588550488802,
        -0.014491137644956758,
        0.01263138507308084,
        -0.019242602763084392,
        -0.0733303381866722,
        -0.018292906049445227,
        0.05203174273215702,
        0.07108568817595774,
        -0.08343963472779074,
        0.0030822278891553183,
        -0.043079822747850896,
        0.0717063084658384,
        -0.1056287845026471,
        -0.06327950593031756,
        -0.03616614610102648,
        0.010053277654244241,
        -0.03152723390422741,
        -0.07269696181397854,
        -0.014079695124384633,
        0.08578750922566855,
        -0.04307267170810249,
        -0.04402825272704562,
        -0.09682720427838318,
        0.06375616718015537,
        -0.04947436756454398,
        -0.0029785530157637892,
        0.03039949058876706,
        -0.03539485514329275,
        -0.05548076965065344,
        -0.024228377309555585,
        0.027550652504633434,
        0.011002345049667904,
        -0.004858191929030838,
        -0.019074241583855886,
        0.027600071417816837,
        0.011654087395898708,
        -0.02631848866248684,
        -0.02578257118869761,
        -0.10142546713137925,
        -0.018439272660266164,
        0.06011861999639479,
        -0.05792629939768642,
        0.007650473160388653,
        0.007216346334508671,
        -0.07229130036291806,
        0.11354801219454128,
        -0.05876490669751221,
        -0.048680575355518445,
        0.03998858324521954,
        -0.03099187589401133,
        -0.07528775844678472,
        0.0807709042464839,
        0.02594194889502386,
        -0.059670671535681064,
        0.00942776604424093,
        -0.0028776227121772493,
        -0.007967111744846798,
        -0.04434327492245102,
        -0.06902357179670565,
        0.026050451976602083,
        -0.02869977520706543,
        -0.0952196352838029,
        -0.025971154170693414,
        -0.0786774800816319,
        0.04761626233747273,
        -0.08901675908698108,
        0.1119433934110706,
        0.030254175512423296,
        0.01059367145737672,
        -0.05670020678971586,
        -0.03394356639743115,
        0.022393842414101854,
        0.025449602137129233,
        0.025939369196329002,
        -0.050713064241734926,
        0.077439325499781,
        0.061545359517473766,
        0.02526940486081912,
        0.03266726812035888,
        -0.08141116738551413,
        -0.09018556395340667,
        0.010151848657573976,
        0.08072930194832235,
        0.0553609112563701,
        0.0246317098519104,
        0.007312979473992716,
        0.04052775692134706,
        0.053557329947192076,
        -0.009912575298792916,
        -4.1162550004678645e-05,
        0.02064051228316809,
        0.06269082956006024,
        0.01436611965035707,
        0.049156134832089554,
        0.05156397795174512,
        -0.05457823351049779,
        0.05043153355322838,
        0.054465384872352425,
        0.0032236866106755326,
        0.044691090793329266,
        -0.01502912518723486,
        -0.05278259613848889,
        -0.04349356329374674,
        -0.042364616259241004,
        0.10373610912461664,
        -0.060369774570427345,
        0.019803705990885937,
        0.0017775956240031058,
        0.014881825685941149,
        0.031176606389590807,
        0.005187161942116694,
        0.11537231978169586,
        0.05962628930302614,
        0.08926838432742581,
        0.003948152046781459,
        0.03947227740380522,
        0.08387826027893495,
        0.04569313728543766,
        -0.004052022963301519,
        -0.09258751393249197,
        -0.0025134750463571202,
        0.016596275882914797,
        -0.020897348849089958,
        0.12446669332718438,
        -0.04581130154901584,
        -0.02715954106065882,
        -0.02016855114462183,
        0.024126115695269356,
        0.022103964738976425,
        0.031706028953720745,
        0.03458119973085572,
        -0.0622546128240966,
        0.017783922774863316,
        -0.0257411088580495,
        0.014671121020176139,
        -0.01566458047634695,
        0.08307653347444649,
        0.021801702382786094,
        0.02288190914923984,
        -0.09569353561874552,
        0.02194727245531278,
        0.012684137735847801,
        -0.013590042903889753,
        -0.022486994229960304,
        0.006931704211008772,
        0.021328056668308015,
        0.014223522570203376,
        0.009979506590761767,
        0.057073022598144924,
        -0.09848105316542527,
        0.043156021274604606,
        0.03524018650437819,
        0.027623847248296796,
        0.08476597955375649,
        0.007239602068100677,
        0.05145304487490161,
        -0.06462512284282648,
        0.006744366563612844,
        0.031447484196620516,
        -0.00821805822485713,
        -0.04280214649968832,
        -0.05164597183518455,
        -0.049775177129415545,
        -0.04296215739967706,
        -0.01273663106337593,
        -0.1092107429319847,
        0.004494659248453175,
        -0.05411978625444536,
        -0.0209674709519419,
        -0.0388417915067658,
        0.04687245792513008,
        -0.0670781976466051,
        0.041695730573428014,
        0.01738314907038482,
        -0.12066013682760883,
        -0.0007278346140906552,
        -0.03599612340010885,
        0.0046232909556884446,
        -0.04301849686707173,
        0.02750734954436447,
        -0.04130865164685524,
        -0.05407817320532628,
        0.01491437475546039,
        0.03261562476361068,
        0.018931905751488322,
        0.018819775757460373,
        0.049097392661951206,
        0.08631393988574831,
        -0.018063878096220478,
        -0.0377538822596216,
        -0.07675479194226553,
        0.024942908036097895,
        0.10160400933616409,
        0.09295558616780743,
        -0.1339290189310165,
        0.14891022223144681,
        0.009061839159743651,
        0.042449503780286085,
        -0.004910539295859614,
        -0.04224893558442232,
        0.05105517634774707,
        -0.11120492246752428,
        0.05924272683503981,
        0.005041409091854501,
        -0.042159187126018106,
        -0.02861482312755103,
        -0.008975354846212352,
        -0.06483237811354399,
        0.04919556340450798,
        0.03263723103874385,
        0.01915652623272444,
        -0.009258200725315788,
        0.03098275206447058,
        0.014819057343427804,
        0.043921910468693544,
        0.06390264564322931,
        -0.0030316503330981718,
        -0.0144763011903826,
        -0.039594161798956626,
        0.02249651527878204,
        -0.04091185523454755,
        0.012781534878280447,
        0.11940998726242226,
        0.020493411204439257,
        0.02109543248232606,
        -0.018945865906889472,
        0.06790050222814285,
        -0.12022800883869364,
        0.007597690262415848,
        0.04322704646182151,
        -0.023172489902907042,
        -0.04269598451062078,
        0.07222037739183126,
        0.02581102034841334,
        -0.08198045124192294,
        0.006383895165706413,
        -0.002973159608344887,
        -0.01468235367070308,
        0.0925119166693521,
        -0.02010647911508766,
        0.07103060726632524,
        0.0050628770113283934,
        -0.13371486433605378,
        0.017545248827192748,
        0.008672376673923376,
        0.03901900817936508,
        0.00028434878370751666,
        0.008441475388955542,
        0.011607983326892345,
        -0.049030159704160726,
        -0.1443763505861614,
        0.013302109575900481,
        0.06036419989666,
        0.02226418067914241,
        -0.07346043790172191,
        -0.013253323864050448,
        -0.0059747572169374175,
        0.0498806837492862,
        0.024515384808384777,
        0.045592370033208256,
        0.043089530377956445,
        -0.07241312005019422,
        -0.0022229657142427903,
        0.0008592395842323218,
        0.016441377983407914,
        0.07505110520920424,
        0.026983276041370314,
        -0.010868407271505254,
        -0.08251224325355253,
        0.006532953128473198,
        0.0928104624322323,
        -0.10769574444828063,
        -0.008717836953690225,
        -0.006969545065013117,
        0.10178318700128301,
        -0.01310732266459529,
        0.009178270982902496,
        0.003687018788569043,
        -0.07699583060742071,
        0.0242989329711667,
        0.13409930560178884,
        -0.042897577817084295,
        -0.024992270722117134,
        -0.007197028275025465,
        0.08355706754535366,
        0.03245154493014019,
        0.10695269188368023,
        0.020554283650146443,
        -0.03417601582306377,
        0.0257124462581086,
        0.02012822281724422,
        -0.057971902584720174,
        0.07338212403532554,
        -0.08141598873050497,
        -0.07825419922843913,
        0.007318613460816974,
        0.08626747437548986,
        0.008975567730751406,
        0.03225142002785613,
        -0.024380241528363985,
        -0.019450854230942654,
        -0.049995608139035946,
        0.010425112803054304,
        -0.024380898681833095,
        -0.008597984574158746,
        0.03011770261452802,
        0.0021162816581700907,
        -0.044187171725595296,
        0.02220113212109878,
        0.0289209576378741,
        -0.05614043197729849,
        0.014960475989143546,
        0.02387428429822321,
        -0.003393649540359895,
        0.005388099923541394,
        -0.05105788880844923,
        0.030972966471782836,
        -0.020363969761000127,
        -0.04270613471205973,
        0.023163646967380446,
        0.040518622340733286,
        -0.019635619589142225,
        -0.014010052417316558,
        0.0030951666320926147,
        0.05172546345131271,
        -0.057056923077536084,
        0.009337586243800568,
        0.012050340429099462,
        0.00934549204819406,
        0.044403862807090785,
        -0.05473750982138884,
        0.032489958837625184,
        -0.06151970415739592,
        0.04186697650857672,
        -0.012557638993465469,
        -0.051339463407887344,
        -0.004136803141949018,
        0.001833830314228449,
        -0.054429353829812954,
        -0.0032352266987476794
      ],
      "metadata": {
        "confidence": 0.9,
        "provenance": "control_system_v2",
        "timestamp": "2025-10-20T03:03:42.442151",
        "logical_form": "update(internal_state) IF phase_lock(144000)",
        "token_span": null,
        "model_id": null,
        "custom": {}
      },
      "is_anchor": false
    },
    {
      "id": "64bf7778-00c2-4d45-9f0a-c0e328e4ffec",
      "type": "surface_text",
      "text": "AI updates internal model when phase lock is detected at 144000 cycles",
      "embedding": [
        0.042494461160047764,
        -0.05065798896362518,
        0.012253440130574342,
        -0.054024858861451495,
        0.041535910018478804,
        -0.02396703864613811,
        -0.005906678283118023,
        0.02926884351000246,
        0.02285723062559538,
        0.024398862841780496,
        0.062100924027460896,
        -0.02232614412832949,
        -0.12092544924818391,
        0.09994600804744544,
        -0.04342644635133131,
        0.05568781815946813,
        -0.04277576175229853,
        0.01139849037842173,
        0.154425037412179,
        0.05059368537193236,
        0.01926892356198234,
        0.012423243206778716,
        0.0921565221480901,
        -0.022885614652981053,
        0.004587053415802551,
        0.021322930172362396,
        -0.02576042494858964,
        -0.06484617682062334,
        0.0871583273487801,
        0.03305183614775865,
        -0.039802969215556955,
        0.015384565049907912,
        0.07369512192438735,
        0.02356419618190115,
        -0.03154070591846332,
        0.008735188848884651,
        0.02379069060193044,
        -0.077529995730049,
        0.008810309410523503,
        0.010652326431885741,
        0.13939376442731854,
        -0.041012576996508794,
        0.02065370513358231,
        0.05675349430944695,
        -0.026222744526151375,
        -0.14503470346656266,
        -0.04551718598922825,
        -0.03737948389729033,
        -0.008227697971068629,
        0.017194766844188924,
        0.01306843283951545,
        0.028526948257500937,
        0.006960539929546068,
        0.03139831577931694,
        0.01709080786491161,
        0.024356626077748782,
        0.046530660068842926,
        -0.05241476614200155,
        -0.04325045944047741,
        0.02499255014763501,
        -0.029383680808737475,
        0.012254492057221199,
        0.1628536135884625,
        0.033745578430043555,
        0.0020813375727279926,
        -0.0008678846202350801,
        0.059488748674099055,
        -0.03555776242160331,
        0.012883313150900013,
        0.0021486955756862655,
        0.08211762791284177,
        0.04994235286833635,
        -0.03391999520494333,
        -0.06334235024218282,
        0.003638532815598478,
        -0.009412649640589969,
        -0.011023998733915038,
        0.01616928643506112,
        -0.020934614058593707,
        0.017973384161736616,
        -0.006684888435595182,
        0.040259361709580326,
        0.010738960935485691,
        -0.03684808676950136,
        -0.05361226437798158,
        -0.001966035597381618,
        0.013837067997356345,
        -0.0013991807277786799,
        0.05889531039836228,
        -0.023690869676569245,
        0.08766113201472285,
        -0.030180494404199995,
        0.0674086587470535,
        0.016582067962020313,
        0.0019742087358573055,
        -0.06606105658432154,
        0.03398378730629584,
        0.05324212675955785,
        0.014751817301029223,
        0.05664179158274987,
        -0.005677538281024488,
        0.06831049757064285,
        0.041304514444719755,
        -0.08183662471536711,
        -0.004906201332377322,
        0.07365498706359623,
        -0.023914568521294418,
        -0.004367230498143042,
        -0.03147271466585449,
        -0.02398146721265583,
        0.036771797149517595,
        0.033449458985385726,
        0.07307329500708307,
        0.008482337345492008,
        -0.011897083580152323,
        0.01655079672394693,
        0.06172489608388066,
        0.01446303980220894,
        0.048898894002946704,
        -0.035965062521409755,
        0.0710467467743977,
        0.023214794243377704,
        -0.0823501426257545,
        -0.07662486916964571,
        -0.08827424478163601,
        0.017499998192426928,
        0.04036193624462058,
        -0.015726713647189288,
        -0.03111824345569749,
        -0.054089868328803255,
        0.0024821089136015054,
        0.0013321128866105873,
        0.11002261736499996,
        -0.06109389886503444,
        0.11104396243899126,
        -0.06396158087068073,
        0.052109212572176745,
        0.02071118343262042,
        0.046706922422519295,
        -0.02801500344126801,
        0.10889870969517626,
        0.027486095006431885,
        -0.11334903917748283,
        -0.018568840260584956,
        0.07751811054219665,
        0.005257751722192783,
        0.021772214677315424,
        -0.035109102277165675,
        0.004448244327583898,
        -0.05861155162003628,
        -0.0011264545825137337,
        0.04214892978814708,
        -0.011915825252078174,
        -0.07155960276273282,
        0.03258993200462426,
        0.08546944141878485,
        -0.03306728707733964,
        0.11497344074168278,
        0.059506062106514525,
        -0.09411823519636099,
        0.05409968043554974,
        -0.004335587885343608,
        -0.01250539234513445,
        0.015828271304816743,
        -0.0028353222044198577,
        -0.037617956680918725,
        0.09394208148869496,
        -0.000569251915355703,
        -0.03834680238985743,
        0.05832064685732282,
        0.09573997162026261,
        0.03659163393747351,
        0.012706650264672277,
        -0.006593914032959572,
        -0.04671419675804993,
        -0.026087995452159375,
        -0.04767749772689856,
        0.12771513726530798,
        -0.04965112765116634,
        0.04726143618118454,
        -0.10026892658842432,
        0.018781574434568522,
        -0.06617253439600575,
        0.03147036569819526,
        -0.03727227872788359,
        -0.0030639460539881097,
        -0.09547840294289071,
        0.1273181440734075,
        -0.048551507858384815,
        0.07892593068418831,
        0.011454786187032294,
        -0.03181278615754983,
        -0.021362841928943863,
        0.018506845622089044,
        -0.004939036968575149,
        -0.047984237430749845,
        0.16788769944325843,
        0.04629075599587603,
        0.009807611791563985,
        0.009897562683295204,
        -0.03683197314172633,
        0.0009656390223443103,
        0.14915488890984607,
        -0.0620434427729146,
        -0.000721897835304841,
        -0.0327327282358321,
        -0.05575285886856514,
        0.06806089024140635,
        0.024210348236379776,
        0.025435378622639254,
        -0.05739794623742354,
        -0.03949495986542194,
        0.04089361061002146,
        0.052907472469746794,
        -0.07287428274213416,
        -0.081007323321459,
        0.02904267017931855,
        0.0046353618274313614,
        -0.05030995715345714,
        -0.051766617632851,
        -0.010278098929803439,
        0.0229467424624723,
        -0.008114485019876005,
        -0.00025586575399356775,
        0.04374097591903027,
        0.010269955261346537,
        -0.01847746181604552,
        0.0385436073691621,
        -0.04561075762448421,
        -0.024711171893451924,
        -0.04794695097718504,
        -0.046393592495846014,
        -0.02385273914888146,
        -0.012943558054798484,
        0.030313965595524735,
        -0.09039299589572261,
        0.07472137424740759,
        0.03132587407086657,
        0.008527555794824587,
        0.03531521377379377,
        -0.04431916514527924,
        -0.002758987216602645,
        0.0293834199633403,
        -0.026681428099138753,
        0.0012996491291057367,
        0.003317232292087158,
        -0.056231866677194944,
        0.04138519281489146,
        -0.07433241152307019,
        0.07407484289997208,
        -0.06597239863404923,
        0.0061162269674214,
        0.039732680532835095,
        0.0171449312600639,
        -0.07454642963818417,
        -0.05461360090158678,
        0.02286271728920322,
        -0.0788253941792138,
        -0.022625248399438636,
        -0.10924497758222745,
        -0.010874274325830024,
        0.0073397591806698475,
        -0.06227674181371803,
        -0.018458156210024147,
        0.026487481150358923,
        -0.00683918067350583,
        -0.030881035934503486,
        0.01934756575787205,
        -0.0472889886551083,
        -0.09641964985113086,
        0.006563954091834968,
        -0.016851126040098086,
        -0.04651395421838241,
        0.03510412573252498,
        0.031960735673254345,
        -0.01580539660544001,
        0.03949214591476465,
        -0.09455455493120023,
        0.08808370505521612,
        0.06524599882215287,
        -0.09441298224369402,
        -0.04019352062863669,
        0.05868455649125559,
        -0.0694986657039855,
        0.09334723343343548,
        -0.07740737530204815,
        -0.044231127204037345,
        0.016483805054207604,
        -0.08262059441616189,
        -0.05971433366364026,
        0.03601496251325034,
        -0.028773756757402902,
        -0.00540698298142019,
        -0.01273392539951537,
        0.004107631829753034,
        -0.0920190863508846,
        0.061484091047045636,
        0.009847570568603192,
        -0.03338139055769652,
        0.04961231434043145,
        0.061617218961956925,
        -0.06031363494352495,
        0.008473932995854947,
        -0.03854723831200084,
        0.004782981719986545,
        0.004971783519148423,
        -0.004534268216207184,
        -0.02919884810267689,
        0.022194738550814167,
        0.004163277060059088,
        0.05502176213209304,
        -0.026596072812966345,
        0.0009073364260634734,
        -0.01640079661649407,
        0.10405584565372165,
        0.021444344178442647,
        -0.09778075232658907,
        -0.005195568515465076,
        0.013828417141293322,
        0.008837083927710923,
        0.04845914820249111,
        0.052702849829278905,
        -0.01014529023136498,
        0.05494838548483041,
        -0.04058466866077445,
        0.05127337652640178,
        0.017363803580968076,
        -0.06328563425510875,
        -0.002435988553314413,
        -0.06493300212573892,
        -0.026829428979914274,
        -0.011212354992769942,
        0.029929176317042817,
        -0.03740715926507682,
        -0.05883321473071094,
        0.028063296116348203,
        -0.043529140149939076,
        0.0077038832654163876,
        -0.008775343267280987,
        -0.06689128045814503,
        0.028052947144648357,
        -0.007280617131921634,
        -0.05043457357164376,
        0.017001316842207296,
        0.04599750216791177,
        0.020141521319033847,
        0.02175873010197184,
        -0.07096269564294018,
        -0.1079033605122535,
        -0.028721626636721478,
        0.01234326902077421,
        -0.02742929723858572,
        -0.043424921945588474,
        0.08542365003733814,
        0.06092772490794237,
        -0.008487467939087816,
        -0.00043813337923427175,
        0.05931069862866315,
        0.05562680342519808,
        0.05627793917488799,
        0.017017550445668113,
        0.014143039750173609,
        0.0011828983297298065,
        0.013736527150209944,
        -0.06628337992106464,
        -0.00548319271883784,
        0.01387560864335546,
        0.045285272408448526,
        -0.04411203755602432,
        -0.0033997374133188803,
        -0.0011008235315276916,
        0.004725916559075407,
        -0.019712732754539098,
        0.042544698166067665,
        -0.019685579163306505,
        -0.029618878602847367,
        -0.06148185243934713,
        0.003392262184877001,
        0.10192909079553437,
        0.031323115864097834,
        0.033598434881073135,
        -0.031589757126597506,
        0.06902559022158905,
        -0.055953904446247864
      ],
      "metadata": {
        "confidence": 0.88,
        "provenance": "reasoning_engine",
        "timestamp": "2025-10-20T03:03:42.442155",
        "logical_form": null,
        "token_span": null,
        "model_id": "claude-sonnet-4-5",
        "custom": {}
      },
      "is_anchor": false
    }
  ],
  "edges": [
    {
      "from_id": "6243f0ef-5fc4-4900-8ee1-146d8625bf77",
      "to_id": "b7fe5a4c-3826-4926-9c8f-6ee16aa352f1",
      "type": "inference",
      "weight": 0.95,
      "rule_id": "observer_perception_link",
      "metadata": {}
    },
    {
      "from_id": "b7fe5a4c-3826-4926-9c8f-6ee16aa352f1",
      "to_id": "9a4ff851-2ef3-4b22-8454-1f6c1e9d0e00",
      "type": "transform",
      "weight": 0.9,
      "rule_id": "FFT_detection",
      "metadata": {}
    },
    {
      "from_id": "9a4ff851-2ef3-4b22-8454-1f6c1e9d0e00",
      "to_id": "99f50fc8-941f-4a94-b2fc-bb9d74d424c2",
      "type": "implication",
      "weight": 0.92,
      "rule_id": "phase_lock_trigger",
      "metadata": {}
    },
    {
      "from_id": "99f50fc8-941f-4a94-b2fc-bb9d74d424c2",
      "to_id": "64bf7778-00c2-4d45-9f0a-c0e328e4ffec",
      "type": "inference",
      "weight": 0.88,
      "rule_id": "explanation_synthesis",
      "metadata": {}
    }
  ],
  "metadata": {
    "node_count": 5,
    "edge_count": 4,
    "timestamp": "2025-10-20T03:03:42.449450"
  }
}

Please peer review this work. I hope to make this public.