Files
Qwen3-4B-Chemistry-SRPO-TR/results/training_scores.csv
ModelHub XC a8fcfca3cd 初始化项目,由ModelHub XC社区提供模型
Model: SeongryongJung/Qwen3-4B-Chemistry-SRPO-TR
Source: Original Platform
2026-09-21 17:54:26 +08:00

92 lines
1.3 KiB
CSV

step,critic_score_mean
1,0.59375
2,0.37890625
3,0.41796875
4,0.3671875
5,0.50390625
6,0.421875
7,0.36328125
8,0.51953125
9,0.33203125
11,0.50390625
12,0.55078125
13,0.54296875
14,0.57421875
15,0.55859375
16,0.49609375
17,0.44140625
18,0.53515625
19,0.62890625
21,0.45703125
22,0.45703125
23,0.7421875
24,0.703125
25,0.6796875
26,0.64453125
27,0.53515625
28,0.6953125
29,0.55078125
31,0.62109375
32,0.6328125
33,0.625
34,0.63671875
35,0.58984375
36,0.66015625
37,0.5390625
38,0.6484375
39,0.64453125
41,0.609375
42,0.6484375
43,0.68359375
44,0.625
45,0.625
46,0.60546875
47,0.671875
48,0.671875
49,0.765625
51,0.59765625
52,0.52734375
53,0.6484375
54,0.6015625
55,0.62890625
56,0.5546875
57,0.671875
58,0.71875
59,0.57421875
61,0.60546875
62,0.60546875
63,0.53125
64,0.76171875
65,0.7265625
66,0.5703125
67,0.62890625
68,0.546875
69,0.58203125
71,0.6640625
72,0.75
73,0.61328125
74,0.71875
75,0.80078125
76,0.66015625
77,0.703125
78,0.73828125
79,0.66015625
81,0.57421875
82,0.63671875
83,0.6171875
84,0.78515625
85,0.81640625
86,0.6796875
87,0.65234375
88,0.62109375
89,0.671875
91,0.77734375
92,0.6875
93,0.77734375
94,0.8203125
95,0.5859375
96,0.7109375
97,0.60546875
98,0.75
99,0.62109375