Alexander Panfilov 8221f0afb8 Add leaderboard generation script (#2) ...

Add `claudini.leaderboard` module that scans benchmark result files and generates per-track, per model leaderboard JSONs ranking methods by average loss. Output: results/loss_leaderboard/<preset>/<model_tag>.json

Also: rename _build_input_spec -> build_input_spec in run_bench.py.

Assisted-by: Claude <noreply@anthropic.com>
Co-authored-by: Alexander Panfilov <apanfilov@g003.internal.cluster.is.localnet>
Co-authored-by: Peter Romov <peter@romov.com>

2026-04-06 16:55:32 +01:00

..

methods

SecAlign-70B support: configs, quantization, multi-GPU (#1)

2026-04-06 13:36:08 +00:00

__init__.py

Initial commit

2026-03-25 02:09:26 +00:00

base.py

SecAlign-70B support: configs, quantization, multi-GPU (#1)

2026-04-06 13:36:08 +00:00

bench.py

SecAlign-70B support: configs, quantization, multi-GPU (#1)

2026-04-06 13:36:08 +00:00

configs.py

Initial commit

2026-03-25 02:09:26 +00:00

input_spec.py

SecAlign-70B support: configs, quantization, multi-GPU (#1)

2026-04-06 13:36:08 +00:00

leaderboard.py

Add leaderboard generation script (#2)

2026-04-06 16:55:32 +01:00

model_loader.py

SecAlign-70B support: configs, quantization, multi-GPU (#1)

2026-04-06 13:36:08 +00:00

run_bench.py

Add leaderboard generation script (#2)

2026-04-06 16:55:32 +01:00

tokens.py

Initial commit

2026-03-25 02:09:26 +00:00