Files
Qwen3-4B-Tooluse-GRPO-TR/results/training_scores.csv
ModelHub XC 9af9efe7e1 初始化项目,由ModelHub XC社区提供模型
Model: SeongryongJung/Qwen3-4B-Tooluse-GRPO-TR
Source: Original Platform
2026-08-16 11:24:56 +08:00

92 lines
1.2 KiB
CSV

step,critic_score_mean
1,0.3125
2,0.24609375
3,0.34375
4,0.2734375
5,0.375
6,0.42578125
7,0.3828125
8,0.3046875
9,0.28515625
11,0.30078125
12,0.359375
13,0.26953125
14,0.25390625
15,0.5390625
16,0.3125
17,0.38671875
18,0.3046875
19,0.23828125
21,0.20703125
22,0.4453125
23,0.20703125
24,0.2890625
25,0.2109375
26,0.5234375
27,0.375
28,0.30078125
29,0.171875
31,0.421875
32,0.28125
33,0.2265625
34,0.359375
35,0.30078125
36,0.4140625
37,0.42578125
38,0.3828125
39,0.42578125
41,0.30859375
42,0.22265625
43,0.3125
44,0.265625
45,0.25390625
46,0.30859375
47,0.30078125
48,0.38671875
49,0.2109375
51,0.37109375
52,0.4375
53,0.36328125
54,0.3359375
55,0.4140625
56,0.36328125
57,0.4296875
58,0.37109375
59,0.4140625
61,0.3828125
62,0.33984375
63,0.3671875
64,0.38671875
65,0.30859375
66,0.421875
67,0.421875
68,0.46484375
69,0.5390625
71,0.3046875
72,0.38671875
73,0.33984375
74,0.421875
75,0.35546875
76,0.25
77,0.30078125
78,0.375
79,0.4140625
81,0.40625
82,0.44140625
83,0.25390625
84,0.46875
85,0.390625
86,0.40234375
87,0.4375
88,0.53515625
89,0.30859375
91,0.47265625
92,0.41015625
93,0.32421875
94,0.3125
95,0.30859375
96,0.42578125
97,0.33984375
98,0.359375
99,0.3828125