File size: 843 Bytes
24918f7 48f76bc | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | # Training from the archived inputs
open1-b7g-e4 derives from open1-b5-e4: the same rows in the same order, with `labels.guides`
relabelled from tool calls (`training/turn-load/relabel_guides.py`). Train the B7 and
B7G heads with `training/turn-load/train.sh`:
- `train.sh GPU B7 RUN`: `--families tools,guides` on `corpora/independent.json`,
`--tool-truth consensus --tool-weight none --tool-negatives 24 --pos-weight 6
--lr 0.0005 --seed 11 --batch-size 64 --epochs 45 --patience 8`.
- `train.sh GPU B7G RUN`: the same with `--families guides`.
Score a head on every tool and guide option with `training/turn-load/turn_probe.py` and
audit omissions with `training/turn-load/guide_audit.py`. The backbone is
`convaiinnovations/laya-multilingual` at revision
`e4e9ddf21a7b1903b7acffd8814ad4307bf63a67`; obtain that revision explicitly.
|