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

1	step	critic_score_mean
2	1	0.5625
3	2	0.53515625
4	3	0.62109375
5	4	0.578125
6	5	0.56640625
7	6	0.640625
8	7	0.6484375
9	8	0.58984375
10	9	0.6328125
11	11	0.4921875
12	12	0.58984375
13	13	0.54296875
14	14	0.76953125
15	15	0.66015625
16	16	0.6953125
17	17	0.5703125
18	18	0.5546875
19	19	0.63671875
20	21	0.546875
21	22	0.75
22	23	0.80859375
23	24	0.89453125
24	25	0.6796875
25	26	0.65234375
26	27	0.8203125
27	28	0.6171875
28	29	0.68359375
29	31	0.63671875
30	32	0.6640625
31	33	0.8203125
32	34	0.7890625
33	35	0.7734375
34	36	0.671875
35	37	0.86328125
36	38	0.6171875
37	39	0.77734375
38	41	0.5859375
39	42	0.68359375
40	43	0.5859375
41	44	0.73828125
42	45	0.78515625
43	46	0.78125
44	47	0.69921875
45	48	0.83984375
46	49	0.640625
47	51	0.703125
48	52	0.76953125
49	53	0.6328125
50	54	0.5546875
51	55	0.703125
52	56	0.81640625
53	57	0.7578125
54	58	0.828125
55	59	0.65234375
56	61	0.81640625
57	62	0.8125
58	63	0.71875
59	64	0.74609375
60	65	0.60546875
61	66	0.70703125
62	67	0.78515625
63	68	0.81640625
64	69	0.83203125
65	71	0.74609375
66	72	0.7734375
67	73	0.70703125
68	74	0.83203125
69	75	0.69140625
70	76	0.734375
71	77	0.5546875
72	78	0.609375
73	79	0.7734375
74	81	0.71875
75	82	0.7265625
76	83	0.6875
77	84	0.7421875
78	85	0.7421875
79	86	0.84765625
80	87	0.625
81	88	0.6640625
82	89	0.74609375
83	91	0.80859375
84	92	0.7109375
85	93	0.78515625
86	94	0.73828125
87	95	0.76171875
88	96	0.58984375
89	97	0.85546875
90	98	0.8828125
91	99	0.71875