consolidate eval scripts to have single eval.py

This commit is contained in:
Andreas Koepf 2025-02-25 16:13:22 +01:00
parent bea806fe3c
commit e7ae82a831
12 changed files with 104 additions and 337 deletions

11
eval/yaml/logic.yaml Normal file
View file

@ -0,0 +1,11 @@
model: deepseek/deepseek-r1
category: logic
datasets:
- propositional_logic
- self_reference
- syllogism
- zebra_puzzles
eval_dir: eval/r1
dataset_size: 50
dataset_seed: 42
developer_role: system