Jakub
67061cf373
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.
7 lines
110 B
JSON
7 lines
110 B
JSON
{
|
|
"_from_model_config": true,
|
|
"bos_token_id": 1,
|
|
"eos_token_id": 2,
|
|
"transformers_version": "5.5.0"
|
|
}
|