Skip to content
No results
  • Explainers
J-Space logo
J-Space
  • Explainers
J-Space logo
J-Space
  • Philosophy of AI

Decoder-Only Transformers Route Tokens Instead of Igniting a Global Workspace

QKV attention routes tokens rather than igniting broadcast. Decoder-only transformers fail GWT; the 2017 paper dropped recurrence. Use indicator tests.

  • Editorial Team
  • September 1, 2026
  • Deep Dives

Claude Residual Streams Need 34 Million SAE Features, Not Chat Logs

Decode residual-stream activations and 34 million SAE features; chat transcripts, chain of thought, and the public API do not expose them.

  • Editorial Team
  • August 31, 2026
  • Philosophy of AI

Neither Recurrent Loops Nor Untied Depth Give Transformers Self-Monitoring

GPT-3 stacks 175 billion untied parameters while Universal Transformers reuse one block. Both broadcast. Neither creates C2 sentience.

  • Editorial Team
  • August 26, 2026
  • Deep Dives

Unembedding Proximity Pulls Residual Effects Into Late Layers

Late residual writes sit immediately upstream of the unembedding and bias decoder-aligned effects toward the last blocks.

  • Editorial Team
  • August 25, 2026
  • Deep Dives

Why Extra Residual Features Fail to Add Workspace Slots

Published widths from 768 to 12288 bound how many residual directions stay independent. Extra features interfere instead of adding workspace slots.

  • Editorial Team
  • August 23, 2026
  • Deep Dives

Five Tests Separate Steerable SAE Features From Reportable Runtime Monitors

Apply five tests so residual-stream directions become human-labelable, probe-readable and verbally usable. A clamp alone cannot monitor a live model.

  • Editorial Team
  • August 22, 2026
  • AI Safety

Score Hidden Objectives Before Compliant Tokens Appear

Score hidden objectives before a compliant token emits. J-Lens reads residual Jacobians as a complementary readout, not an alignment proof.

  • Editorial Team
  • August 18, 2026
  • Deep Dives

Confirm Layer Selection With Patching Before You Trust Attention Maps

Attention weights leave the real selector hidden. State a claim, then patch activations to show which residual hypotheses a layer amplifies or suppresses.

  • Editorial Team
  • August 17, 2026
  • J-Space

What the J Lens Protocol Isolates That a Logit Lens Cannot

Isolate causal paths and confirm necessity on residual streams with the J Lens joint lens and first-order Jacobian stack.

  • Editorial Team
  • August 15, 2026
  • Deep Dives

Tracing Causal Paths in Claude’s Addition and Safety Circuits

Researchers map how features causally influence Claude outputs by reconstructing prompt specific graphs with sparse autoencoders and cross layer…

  • Editorial Team
  • August 14, 2026
Prev
1 2 3 4
Next
Copyright © 2026 - WordPress Theme by CreativeThemes