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.
2 lines
81 B
Plaintext
2 lines
81 B
Plaintext
arc_sk,hellaswag_sk,m_mmlu_sk,truthfulqa_sk_mc1,truthfulqa_sk_mc2,sklegal,skquad
|