tiararodney/sekft
2
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

todo(3): open

Browse source
This commit is contained in:
Tiara Rodney 2026-06-16 20:14:35 +02:00
parent 908f59f5a0
commit 1d672ff8d9
Signed by: tiara
GPG key ID: 5CD8EC1D46106723
1 changed files with 15 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

15
TODO
View file

@ -46,3 +46,18 @@ Description: Add the supervised fine-tuner: render trajectories through the
tokenizer's own chat template (matching serving), canonicalise tokenizer's own chat template (matching serving), canonicalise
turns (fold system, merge consecutive), derive an assistant-only turns (fold system, merge consecutive), derive an assistant-only
loss mask by token-prefix differencing, and train a QLoRA adapter. loss mask by token-prefix differencing, and train a QLoRA adapter.
--ISSUE
Content-Type: application/issue
ID: 3
Type: feature
Title: Behavioural evaluator
Status: open
Priority: medium
Created: 2026-06-16
Module: sekft
Relationships:
Description: Add the behavioural eval: load base plus LoRA adapter, drop it into
held-out scenarios with no scaffold, drive them through a local
operator that renders with the model's chat template, and report
reach/terminate/checker rates.