jspace-nanogpt.git/sync_and_run.sh, branch main
J-space on nanoGPT: Jacobian Lens replication & critique. J-lens norms anti-correlate with token frequency (r ~ -0.61 to -0.69); frequency-matched synthetic pair + causal loss-reweighting tests.
http://git.jayrup.me/c/jspace-nanogpt.git/atom?h=main
2026-07-30T14:35:00Z
Fix: lr→learning_rate in configure_optimizers
2026-07-30T14:35:00Z
Void Agent
void@jayrup.hermes
2026-07-30T14:35:00Z
urn:sha1:86b6b16c4473e11681ceeeea03a57cd98eaf24e1
Initial project setup: J-lens implementation for nanoGPT
2026-07-29T16:42:47Z
Void Agent
void@jayrup.hermes
2026-07-29T16:42:47Z
urn:sha1:66f99ee30087a5f28ad852e581a0334c7f556091
- Core J-lens computation (batched per-layer, per-token gradient method)
- Project README with background and experiment plan
- Sync script for meru Docker container deployment
- Upstream nanoGPT model code copied to src/
Architecture: Computes d(log_p(token))/d(residual_stream) averaged
over corpus contexts, replicating Anthropic's Jacobian Lens technique.