Files
Qwen3-8B-Material-GRPO-TR/results/training_scores.csv
ModelHub XC aebb065570 初始化项目,由ModelHub XC社区提供模型
Model: SeongryongJung/Qwen3-8B-Material-GRPO-TR
Source: Original Platform
2026-08-07 11:03:18 +08:00

92 lines
1.2 KiB
CSV

step,critic_score_mean
1,0.546875
2,0.421875
3,0.48828125
4,0.5
5,0.54296875
6,0.48828125
7,0.55078125
8,0.54296875
9,0.546875
11,0.50390625
12,0.60546875
13,0.546875
14,0.77734375
15,0.62109375
16,0.640625
17,0.56640625
18,0.6171875
19,0.62109375
21,0.56640625
22,0.671875
23,0.70703125
24,0.828125
25,0.63671875
26,0.734375
27,0.80078125
28,0.5859375
29,0.65625
31,0.625
32,0.61328125
33,0.76171875
34,0.79296875
35,0.76171875
36,0.59765625
37,0.796875
38,0.59765625
39,0.7265625
41,0.56640625
42,0.625
43,0.53515625
44,0.703125
45,0.7890625
46,0.71484375
47,0.65234375
48,0.85546875
49,0.6484375
51,0.67578125
52,0.765625
53,0.640625
54,0.54296875
55,0.75
56,0.78515625
57,0.7265625
58,0.796875
59,0.71484375
61,0.8046875
62,0.8203125
63,0.71484375
64,0.80859375
65,0.66015625
66,0.69140625
67,0.78125
68,0.7734375
69,0.82421875
71,0.7734375
72,0.796875
73,0.71484375
74,0.8359375
75,0.6796875
76,0.734375
77,0.53125
78,0.68359375
79,0.7734375
81,0.703125
82,0.72265625
83,0.73828125
84,0.83203125
85,0.70703125
86,0.83203125
87,0.59375
88,0.59765625
89,0.71875
91,0.84765625
92,0.72265625
93,0.7421875
94,0.68359375
95,0.6953125
96,0.55078125
97,0.83984375
98,0.9296875
99,0.76171875