Files
Qwen3-4B-Material-GRPO-TR/results/training_scores.csv
ModelHub XC 5a9933097b 初始化项目,由ModelHub XC社区提供模型
Model: SeongryongJung/Qwen3-4B-Material-GRPO-TR
Source: Original Platform
2026-08-04 08:21:18 +08:00

92 lines
1.3 KiB
CSV

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