Files
Qwen3-4B-Chemistry-SDPO-TR/results/training_scores.csv
ModelHub XC f0b01151ab 初始化项目,由ModelHub XC社区提供模型
Model: SeongryongJung/Qwen3-4B-Chemistry-SDPO-TR
Source: Original Platform
2026-10-03 22:43:24 +08:00

92 lines
1.3 KiB
CSV

step,critic_score_mean
1,0.59765625
2,0.3828125
3,0.4296875
4,0.33203125
5,0.4609375
6,0.50390625
7,0.4453125
8,0.57421875
9,0.3515625
11,0.45703125
12,0.62109375
13,0.6171875
14,0.53125
15,0.65625
16,0.4765625
17,0.41796875
18,0.515625
19,0.54296875
21,0.5703125
22,0.6015625
23,0.70703125
24,0.67578125
25,0.5703125
26,0.67578125
27,0.62109375
28,0.63671875
29,0.44921875
31,0.62109375
32,0.6015625
33,0.67578125
34,0.6640625
35,0.6015625
36,0.73046875
37,0.609375
38,0.6484375
39,0.61328125
41,0.5625
42,0.6875
43,0.71484375
44,0.6171875
45,0.5234375
46,0.67578125
47,0.5859375
48,0.66796875
49,0.73828125
51,0.625
52,0.5078125
53,0.6640625
54,0.51171875
55,0.7421875
56,0.58984375
57,0.671875
58,0.76171875
59,0.63671875
61,0.69140625
62,0.7109375
63,0.58984375
64,0.84375
65,0.7578125
66,0.59375
67,0.59765625
68,0.609375
69,0.60546875
71,0.59375
72,0.70703125
73,0.61328125
74,0.73046875
75,0.71484375
76,0.5625
77,0.63671875
78,0.69140625
79,0.63671875
81,0.609375
82,0.67578125
83,0.640625
84,0.7578125
85,0.7578125
86,0.5546875
87,0.734375
88,0.6640625
89,0.640625
91,0.671875
92,0.7265625
93,0.7890625
94,0.828125
95,0.7578125
96,0.671875
97,0.6640625
98,0.70703125
99,0.68359375