Qwen3-4B-Material-RLSD-TR/results/training_scores.csv

1	step	critic_score_mean
2	1	0.55078125
3	2	0.5078125
4	3	0.6171875
5	4	0.62109375
6	5	0.55078125
7	6	0.625
8	7	0.66796875
9	8	0.62109375
10	9	0.62109375
11	11	0.48046875
12	12	0.61328125
13	13	0.51953125
14	14	0.7890625
15	15	0.671875
16	16	0.70703125
17	17	0.578125
18	18	0.609375
19	19	0.65625
20	21	0.59375
21	22	0.7109375
22	23	0.78125
23	24	0.875
24	25	0.64453125
25	26	0.66796875
26	27	0.7890625
27	28	0.63671875
28	29	0.6640625
29	31	0.6328125
30	32	0.6875
31	33	0.74609375
32	34	0.8125
33	35	0.765625
34	36	0.66796875
35	37	0.8359375
36	38	0.5859375
37	39	0.75390625
38	41	0.5859375
39	42	0.69140625
40	43	0.546875
41	44	0.734375
42	45	0.8125
43	46	0.765625
44	47	0.6953125
45	48	0.8203125
46	49	0.6171875
47	51	0.671875
48	52	0.7421875
49	53	0.65234375
50	54	0.5703125
51	55	0.73828125
52	56	0.7890625
53	57	0.7578125
54	58	0.8359375
55	59	0.70703125
56	61	0.83984375
57	62	0.82421875
58	63	0.68359375
59	64	0.7890625
60	65	0.62890625
61	66	0.72265625
62	67	0.80078125
63	68	0.8203125
64	69	0.828125
65	71	0.78515625
66	72	0.7890625
67	73	0.6875
68	74	0.84375
69	75	0.66015625
70	76	0.69140625
71	77	0.5
72	78	0.6328125
73	79	0.80078125
74	81	0.69921875
75	82	0.7109375
76	83	0.6953125
77	84	0.84375
78	85	0.71875
79	86	0.81640625
80	87	0.62109375
81	88	0.6640625
82	89	0.73828125
83	91	0.75
84	92	0.69140625
85	93	0.7734375
86	94	0.75
87	95	0.71875
88	96	0.6484375
89	97	0.87890625
90	98	0.921875
91	99	0.7578125