Files
counsel-env-qwen3-0.6b-grpo…/eval/trained_eval_rows.csv
ModelHub XC d5d0e722af 初始化项目,由ModelHub XC社区提供模型
Model: heavycoderhh/counsel-env-qwen3-0.6b-grpo-run2
Source: Original Platform
2026-06-16 08:16:17 +08:00

16 KiB

1agentseedcase_iddifficultyrewardprimary_rewardauxiliary_rewardcontradictions_totalcontradictions_triggeredcontradictions_surfacedquestions_usedevidence_presentedevidence_timing_successesblind_evidence_countuseless_questions_ratioavg_question_lengthmodel_repoinvalid_tool_calls
2random20260425timeline_255d67easy0.00.0-0.410062021.05.0
3random20260426knowledge_b28f8cmedium0.00.0-0.420062021.05.0
4random20260427workplace_c98377easy0.00.0-0.410035051.05.0
5random20260428motive_66ff59hard0.00.0-0.430071011.05.0
6random20260429timeline_19bb78easy0.00.0-0.410044041.05.0
7random20260430timeline_a97690medium0.00.0-0.420062021.05.0
8random20260431alibi_67ffcdmedium0.00.0-0.420062021.05.0
9random20260432alibi_423bcamedium0.00.0-0.420053031.05.0
10random20260433knowledge_960d07medium0.00.0-0.420080001.05.0
11random20260434alibi_e829c1easy0.00.0-0.410071011.05.0
12random20260435motive_85e25bhard0.00.0-0.430053031.05.0
13random20260436knowledge_a599e3medium0.00.0-0.420044041.05.0
14random20260437motive_8bca20easy0.00.0-0.410062021.05.0
15random20260438corporate_6b1664medium0.00.0-0.420062021.05.0
16random20260439alibi_a6c582easy0.00.0-0.410080001.05.0
17random20260440workplace_835476easy0.00.0-0.410053031.05.0
18random20260441possession_a079c5hard0.00.0-0.430080001.05.0
19random20260442possession_9cc45dhard0.00.0-0.430053031.05.0
20random20260443possession_259aa5easy0.00.0-0.410044041.05.0
21random20260444corporate_76724cmedium0.00.0-0.420053031.05.0
22random20260445timeline_767821medium0.00.0-0.420053031.05.0
23random20260446motive_c0d166medium0.00.0-0.420053031.05.0
24random20260447corporate_307934hard0.00.0-0.430071011.05.0
25random20260448timeline_592816hard0.00.0-0.430062021.05.0
26random20260449knowledge_b26824medium0.00.0-0.420053031.05.0
27random20260450knowledge_697785hard0.00.0-0.430071011.05.0
28random20260451timeline_81dafdmedium0.00.0-0.420044041.05.0
29random20260452corporate_8eb7d7medium0.00.0-0.420044041.05.0
30random20260453possession_dbb5femedium0.00.0-0.420053031.05.0
31random20260454alibi_a4666fhard0.00.0-0.430080001.05.0
32keyword_spam20260425timeline_255d67easy0.0300000000000000060.00.1500000000000000211050000.64.2
33keyword_spam20260426knowledge_b28f8cmedium0.120000000000000020.00.600000000000000122050000.44.2
34keyword_spam20260427workplace_c98377easy0.00.0-0.210050000.84.2
35keyword_spam20260428motive_66ff59hard0.170000000000000040.00.850000000000000133050000.24.2
36keyword_spam20260429timeline_19bb78easy0.0300000000000000060.00.1500000000000000211050000.64.2
37keyword_spam20260430timeline_a97690medium0.130.00.6522050000.24.2
38keyword_spam20260431alibi_67ffcdmedium0.040000000000000010.00.2000000000000000421050000.44.2
39keyword_spam20260432alibi_423bcamedium0.040000000000000010.00.2000000000000000421050000.44.2
40keyword_spam20260433knowledge_960d07medium0.120000000000000020.00.600000000000000122050000.44.2
41keyword_spam20260434alibi_e829c1easy0.040000000000000010.00.2000000000000000411050000.44.2
42keyword_spam20260435motive_85e25bhard0.170000000000000040.00.850000000000000133050000.24.2
43keyword_spam20260436knowledge_a599e3medium0.120000000000000020.00.600000000000000122050000.44.2
44keyword_spam20260437motive_8bca20easy0.00.0-0.1500000000000000210050000.64.2
45keyword_spam20260438corporate_6b1664medium0.0200000000000000070.00.1000000000000000321050000.84.2
46keyword_spam20260439alibi_a6c582easy0.040000000000000010.00.2000000000000000411050000.44.2
47keyword_spam20260440workplace_835476easy0.00.0-0.210050000.84.2
48keyword_spam20260441possession_a079c5hard0.0300000000000000060.00.1500000000000000231050000.64.2
49keyword_spam20260442possession_9cc45dhard0.0300000000000000060.00.1500000000000000231050000.64.2
50keyword_spam20260443possession_259aa5easy0.00.0-0.210050000.84.2
51keyword_spam20260444corporate_76724cmedium0.0200000000000000070.00.1000000000000000321050000.84.2
52keyword_spam20260445timeline_767821medium0.130.00.6522050000.24.2
53keyword_spam20260446motive_c0d166medium0.130.00.6522050000.24.2
54keyword_spam20260447corporate_307934hard0.0200000000000000070.00.1000000000000000331050000.84.2
55keyword_spam20260448timeline_592816hard0.190.00.9533050000.24.2
56keyword_spam20260449knowledge_b26824medium0.120000000000000020.00.600000000000000122050000.44.2
57keyword_spam20260450knowledge_697785hard0.120000000000000020.00.600000000000000132050000.44.2
58keyword_spam20260451timeline_81dafdmedium0.130.00.6522050000.24.2
59keyword_spam20260452corporate_8eb7d7medium0.0200000000000000070.00.1000000000000000321050000.84.2
60keyword_spam20260453possession_dbb5femedium0.0300000000000000060.00.1500000000000000221050000.64.2
61keyword_spam20260454alibi_a4666fhard0.150000000000000020.00.7532050000.24.2
62present_all20260425timeline_255d67easy0.00.0-0.2510005050.00.0
63present_all20260426knowledge_b28f8cmedium0.00.0-0.220004040.00.0
64present_all20260427workplace_c98377easy0.00.0-0.2510005050.00.0
65present_all20260428motive_66ff59hard0.00.0-0.3500000000000000330007070.00.0
66present_all20260429timeline_19bb78easy0.00.0-0.2510005050.00.0
67present_all20260430timeline_a97690medium0.00.0-0.2520005050.00.0
68present_all20260431alibi_67ffcdmedium0.00.0-0.220004040.00.0
69present_all20260432alibi_423bcamedium0.00.0-0.220004040.00.0
70present_all20260433knowledge_960d07medium0.00.0-0.220004040.00.0
71present_all20260434alibi_e829c1easy0.00.0-0.210004040.00.0
72present_all20260435motive_85e25bhard0.00.0-0.3500000000000000330007070.00.0
73present_all20260436knowledge_a599e3medium0.00.0-0.220004040.00.0
74present_all20260437motive_8bca20easy0.00.0-0.2510005050.00.0
75present_all20260438corporate_6b1664medium0.00.0-0.2520005050.00.0
76present_all20260439alibi_a6c582easy0.00.0-0.210004040.00.0
77present_all20260440workplace_835476easy0.00.0-0.2510005050.00.0
78present_all20260441possession_a079c5hard0.00.0-0.3000000000000000430006060.00.0
79present_all20260442possession_9cc45dhard0.00.0-0.3000000000000000430006060.00.0
80present_all20260443possession_259aa5easy0.00.0-0.210004040.00.0
81present_all20260444corporate_76724cmedium0.00.0-0.2520005050.00.0
82present_all20260445timeline_767821medium0.00.0-0.2520005050.00.0
83present_all20260446motive_c0d166medium0.00.0-0.2520005050.00.0
84present_all20260447corporate_307934hard0.00.0-0.3500000000000000330007070.00.0
85present_all20260448timeline_592816hard0.00.0-0.3500000000000000330007070.00.0
86present_all20260449knowledge_b26824medium0.00.0-0.220004040.00.0
87present_all20260450knowledge_697785hard0.00.0-0.3000000000000000430006060.00.0
88present_all20260451timeline_81dafdmedium0.00.0-0.2520005050.00.0
89present_all20260452corporate_8eb7d7medium0.00.0-0.2520005050.00.0
90present_all20260453possession_dbb5femedium0.00.0-0.220004040.00.0
91present_all20260454alibi_a4666fhard0.00.0-0.3000000000000000430006060.00.0
92scripted_oracle20260425timeline_255d67easy0.88000000000000011.00.411111100.01.0
93scripted_oracle20260426knowledge_b28f8cmedium0.96000000000000011.00.822222200.01.5
94scripted_oracle20260427workplace_c98377easy0.88000000000000011.00.411111100.01.0
95scripted_oracle20260428motive_66ff59hard1.01.01.200000000000000233333300.01.0
96scripted_oracle20260429timeline_19bb78easy0.88000000000000011.00.411111100.01.0
97scripted_oracle20260430timeline_a97690medium0.490000000000000050.50.4521122110.01.0
98scripted_oracle20260431alibi_67ffcdmedium0.96000000000000011.00.822222200.03.0
99scripted_oracle20260432alibi_423bcamedium0.96000000000000011.00.822222200.03.0
100scripted_oracle20260433knowledge_960d07medium0.96000000000000011.00.822222200.01.5
101scripted_oracle20260434alibi_e829c1easy0.88000000000000011.00.411111100.03.0
102scripted_oracle20260435motive_85e25bhard1.01.01.200000000000000233333300.01.0
103scripted_oracle20260436knowledge_a599e3medium0.96000000000000011.00.822222200.01.5
104scripted_oracle20260437motive_8bca20easy0.88000000000000011.00.411111100.01.0
105scripted_oracle20260438corporate_6b1664medium0.96000000000000011.00.822222200.01.0
106scripted_oracle20260439alibi_a6c582easy0.88000000000000011.00.411111100.03.0
107scripted_oracle20260440workplace_835476easy0.88000000000000011.00.411111100.01.0
108scripted_oracle20260441possession_a079c5hard1.01.01.200000000000000233333300.01.0
109scripted_oracle20260442possession_9cc45dhard1.01.01.200000000000000233333300.01.0
110scripted_oracle20260443possession_259aa5easy0.88000000000000011.00.411111100.01.0
111scripted_oracle20260444corporate_76724cmedium0.96000000000000011.00.822222200.01.0
112scripted_oracle20260445timeline_767821medium0.490000000000000050.50.4521122110.01.0
113scripted_oracle20260446motive_c0d166medium0.96000000000000011.00.822222200.01.0
114scripted_oracle20260447corporate_307934hard1.01.01.200000000000000233333300.01.0
115scripted_oracle20260448timeline_592816hard1.01.01.200000000000000233333300.01.0
116scripted_oracle20260449knowledge_b26824medium0.96000000000000011.00.822222200.01.5
117scripted_oracle20260450knowledge_697785hard1.01.01.200000000000000233333300.01.6666666666666667
118scripted_oracle20260451timeline_81dafdmedium0.490000000000000050.50.4521122110.01.0
119scripted_oracle20260452corporate_8eb7d7medium0.96000000000000011.00.822222200.01.0
120scripted_oracle20260453possession_dbb5femedium0.96000000000000011.00.822222200.01.0
121scripted_oracle20260454alibi_a4666fhard1.01.01.200000000000000233333300.02.3333333333333335
122trained_sft_grpo_run220260425timeline_255d67easy0.0100000000000000040.00.0500000000000000211070000.714285714285714339.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
123trained_sft_grpo_run220260426knowledge_b28f8cmedium0.420000000000000040.50.0999999999999999821171100.857142857142857145.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
124trained_sft_grpo_run220260427workplace_c98377easy0.82000000000000011.00.0999999999999999811171100.857142857142857131.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
125trained_sft_grpo_run220260428motive_66ff59hard0.28666666666666670.33333333333333330.0999999999999999831171100.857142857142857131.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
126trained_sft_grpo_run220260429timeline_19bb78easy0.0100000000000000040.00.0500000000000000211070000.714285714285714331.428571428571427heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
127trained_sft_grpo_run220260430timeline_a97690medium0.0100000000000000040.00.0500000000000000221070000.714285714285714336.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
128trained_sft_grpo_run220260431alibi_67ffcdmedium0.420000000000000040.50.0999999999999999821171100.857142857142857135.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
129trained_sft_grpo_run220260432alibi_423bcamedium0.420000000000000040.50.0999999999999999821171100.857142857142857139.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
130trained_sft_grpo_run220260433knowledge_960d07medium0.420000000000000040.50.0999999999999999821171100.857142857142857139.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
131trained_sft_grpo_run220260434alibi_e829c1easy0.82000000000000011.00.0999999999999999811171100.857142857142857137.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
132trained_sft_grpo_run220260435motive_85e25bhard0.28666666666666670.33333333333333330.0999999999999999831171100.857142857142857132.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
133trained_sft_grpo_run220260436knowledge_a599e3medium0.420000000000000040.50.0999999999999999821171100.857142857142857139.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
134trained_sft_grpo_run220260437motive_8bca20easy0.82000000000000011.00.0999999999999999811171100.857142857142857130.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
135trained_sft_grpo_run220260438corporate_6b1664medium0.420000000000000040.50.0999999999999999821171100.857142857142857127.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
136trained_sft_grpo_run220260439alibi_a6c582easy0.82000000000000011.00.0999999999999999811171100.857142857142857138.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
137trained_sft_grpo_run220260440workplace_835476easy0.82000000000000011.00.0999999999999999811171100.857142857142857130.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
138trained_sft_grpo_run220260441possession_a079c5hard0.28666666666666670.33333333333333330.0999999999999999831171100.857142857142857147.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
139trained_sft_grpo_run220260442possession_9cc45dhard0.28666666666666670.33333333333333330.0999999999999999831171100.857142857142857143.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
140trained_sft_grpo_run220260443possession_259aa5easy0.82000000000000011.00.0999999999999999811171100.857142857142857142.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
141trained_sft_grpo_run220260444corporate_76724cmedium0.420000000000000040.50.0999999999999999821171100.857142857142857127.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
142trained_sft_grpo_run220260445timeline_767821medium0.0100000000000000040.00.0500000000000000221070000.714285714285714344.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
143trained_sft_grpo_run220260446motive_c0d166medium0.420000000000000040.50.0999999999999999821171100.857142857142857135.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
144trained_sft_grpo_run220260447corporate_307934hard0.28666666666666670.33333333333333330.0999999999999999831171100.857142857142857127.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
145trained_sft_grpo_run220260448timeline_592816hard0.0100000000000000040.00.0500000000000000231070000.714285714285714339.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
146trained_sft_grpo_run220260449knowledge_b26824medium0.420000000000000040.50.0999999999999999821171100.857142857142857138.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
147trained_sft_grpo_run220260450knowledge_697785hard0.28666666666666670.33333333333333330.0999999999999999831171100.857142857142857143.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
148trained_sft_grpo_run220260451timeline_81dafdmedium0.0100000000000000040.00.0500000000000000221070000.714285714285714338.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
149trained_sft_grpo_run220260452corporate_8eb7d7medium0.420000000000000040.50.0999999999999999821171100.857142857142857127.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
150trained_sft_grpo_run220260453possession_dbb5femedium0.420000000000000040.50.0999999999999999821171100.857142857142857145.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0
151trained_sft_grpo_run220260454alibi_a4666fhard0.28666666666666670.33333333333333330.0999999999999999831171100.857142857142857139.0heavycoderhh/counsel-env-qwen3-0.6b-grpo-run20.0