qwen3-8b-biology-grpo/eval_mean16.csv

1	step	mean16
2	10	0.34375
3	20	0.40125
4	30	0.5325
5	40	0.55625
6	50	0.5675
7	60	0.5775
8	70	0.55875
9	80	0.52
10	90	0.54375
11	100	0.565