lars 1e7d7d7efd analyze: show model/training params on rendered figures ...

`giant rollout` now records the checkpoint's architecture (mode,
hidden_dim, n_blocks, emb_dim, dropout, conditioning) plus training_epoch
and best_val_loss in its YAML sidecar, using data already loaded from the
checkpoint. condor.py carries those through run_meta, and render.py passes
them to plotstyle's new_figure(params=...) so every plot's subtitle shows
what produced it.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

2026-07-24 09:54:21 +02:00

..

__init__.py

analyze: drive prep/submit from the rollout YAML sidecar

2026-07-23 17:56:36 +02:00

catalog.py

Rewrite analysis as streaming rollout-vs-reference plotting pipeline

2026-07-23 17:38:09 +02:00

condor.py

analyze: show model/training params on rendered figures

2026-07-24 09:54:21 +02:00

context.py

Rewrite analysis as streaming rollout-vs-reference plotting pipeline

2026-07-23 17:38:09 +02:00

grouping.py

Rewrite analysis as streaming rollout-vs-reference plotting pipeline

2026-07-23 17:38:09 +02:00

reduce.py

Rewrite analysis as streaming rollout-vs-reference plotting pipeline

2026-07-23 17:38:09 +02:00

reduced.py

Rewrite analysis as streaming rollout-vs-reference plotting pipeline

2026-07-23 17:38:09 +02:00

render.py

analyze: show model/training params on rendered figures

2026-07-24 09:54:21 +02:00

sources.py

analyze: drive prep/submit from the rollout YAML sidecar

2026-07-23 17:56:36 +02:00

variables.py

Rewrite analysis as streaming rollout-vs-reference plotting pipeline

2026-07-23 17:38:09 +02:00