Jakub 67061cf373 Add Aya 300k learning-rate study and benchmark results ... Document six completed LR runs as preparation for full Aya training on Perun. Record selection of LR 0.00010 by lowest final validation loss, selected-model lm-eval and partial SK-bench results, reproducible provenance, and dataset-access/context-truncation limitations. Preserve existing 30k and Titan material.		2026-10-25 00:33:24 +02:00
..
dataset_access_report.json	Add Aya 300k learning-rate study and benchmark results	2026-10-25 00:33:24 +02:00
export_lr_0p00010.yaml	Add Aya 300k learning-rate study and benchmark results	2026-10-25 00:33:24 +02:00
model_config.json	Add Aya 300k learning-rate study and benchmark results	2026-10-25 00:33:24 +02:00
omitted_tasks.txt	Add Aya 300k learning-rate study and benchmark results	2026-10-25 00:33:24 +02:00
task_configs.json	Add Aya 300k learning-rate study and benchmark results	2026-10-25 00:33:24 +02:00
tasks_sk_zero_available_without_nli.txt	Add Aya 300k learning-rate study and benchmark results	2026-10-25 00:33:24 +02:00
tasks_sk_zero_available.txt	Add Aya 300k learning-rate study and benchmark results	2026-10-25 00:33:24 +02:00
tasks_sk_zero_unavailable.txt	Add Aya 300k learning-rate study and benchmark results	2026-10-25 00:33:24 +02:00
tasks_sk_zero.txt	Add Aya 300k learning-rate study and benchmark results	2026-10-25 00:33:24 +02:00
tasks.txt	Add Aya 300k learning-rate study and benchmark results	2026-10-25 00:33:24 +02:00