
The agent that finally thought in a loop
The agent cycle ran end to end for the first time — after three integration failures that died in silence and a training setup that learned to measure two runs from the same starting point.
5 posts

The agent cycle ran end to end for the first time — after three integration failures that died in silence and a training setup that learned to measure two runs from the same starting point.

The two meters that evaluated format and internal dynamics in TatuEngine lived on a disk that stopped existing during the migration. Training kept running on perplexity alone — until the ruler was rebuilt and returned an uncomfortable number: format hit 100%, accuracy never left the noise.

Run 2 proved that TatuEngine learned Portuguese and forgot English. The next two runs proved something else: I was measuring that forgetting on a sample whose error was larger than the effect — 16 rows out of 166.

After learning to infer (252× GPU speedup), TatuEngine went for the Master-Apprentice cycle: a Qwen2.5-3B teacher with LoRA distilling structured reasoning ([THOUGHT]/[ANSWER]) into the BitMamba-2 1B student. Six versions in one day (v0.17→v0.22.2), 381 examples, and the convergence that wouldn't come until the system message fix.

The saga of making Student BitMamba-1B learn without exploding: Adafactor vs AdamW, saving 4 GB of VRAM, and the 4 bugs that produced NaN during bf16 training.