Swap 'spider' for 'ant' in Claude's mid-layers and 8 legs becomes 6. Anthropic's global workspace paper, the J-lens readout, eval-awareness detection, and its limits.
RL gains sit at 40-60% depth: on Qwen3-8B, training only layer 16 beats full-parameter RL (67.1 vs 66.5). Notes on arXiv 2607.01232 and what it doesn't claim about efficiency.
Measured on M1 Max ComfyUI: QIE 2511's pixel shift comes from the encode node's forced 1MP rescale. Stock ReferenceLatent fixes it; expression, outfit and pose diffs tested.
Traced Gemini's "Do NOT issue search queries to the google search tool" message to a leaked system prompt, plus fixes: new chat, forced search, Saved Info.
Tested Irodori-TTS 500M-v3 on a 4GB RTX 3050 Ti Laptop (Windows): default-voice and zero-shot clone samples, real timings, and the FFmpeg DLL fix for MP3 references.
1,560 edit pairs mined from Claude Code logs, 622 posts scanned with Sonnet, ModernBERT-ja-130m trained on a 16GB M4 Mac mini: zero false positives, 60% recall.
Three LLMs converted the same 10 Japanese scene briefs into Anima (Qwen-DiT) prompts, generated as 60 fixed-seed images on an M1 Max with a merged 3-character LoRA. The Qwen-to-Qwen affinity hypothesis did not survive; a strict formatter brief with character-count locks is what actually moved the results, and two failure modes survive any prompt.
Tested on an M1 Max, NumPy only: Qwen maps a prompt to a JSON of knobs, and a 2D Kuramoto oscillator field renders it. No objects, but composition, color, and motion change with the prompt.
Un-0 swaps neural-net weighted sums for Kuramoto coupled-oscillator physics, hitting FID 6.74 on ImageNet-64. Still GPU-simulated, and the 1000x energy claim is unproven — no chip yet.
Fujitsu's PHOTON claims up to 475x over Transformers, but that's tokens/s/GiB (multi-query memory throughput), not faster single responses. What the 1.2B paper tables, the quality drop, and 9-query integration really show.
Merged kei, kana, and koharu into a single Anima (Qwen-DiT) LoRA and ran my first training on Blackwell (RTX 5090, sm_120). Hands-on log: the cu128 / torch2.8 / SDPA stack swap from the 4090, why the weakest character gets absorbed (caption asymmetry, not rank), and how trigger-only prompts separate three close-packed characters at ep143 without ControlNet.
Tested local Wan video gen on a Radeon 8060S (Strix Halo, 48GB UMA, Windows). ZLUDA can't run stock PyTorch; AMD's TheRock gfx1151 wheel gives native ROCm. FastWan 1.3B in 4min, Wan 14B I2V in 13.6min — VAE decode and 16GB-RAM Segfaults are the real limits.