buzzcode2007/LLM-Self-Detection-Research
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

add: output results for various models and datasets

Browse source
This commit is contained in:
buzz-lightsnack-2007 2024-12-07 23:45:28 +08:00
parent a028413832
commit 39516acf9a
3 changed files with 39 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

13
tests/outputs/results (CoT).CSV Normal file
View file

@ -0,0 +1,13 @@
,model name,dataname,expected,human,AI,unsure,invalid
0,Microsoft Phi,LDAG,human,13,14,12,20
1,Microsoft Phi,HLTC,human,12,11,19,17
2,Microsoft Phi,LDAG,AI,10,18,14,17
3,Microsoft Phi,HLTC,AI,14,13,23,9
4,Gemma 2,LDAG,human,35,24,0,0
5,Gemma 2,HLTC,human,45,14,0,0
6,Gemma 2,LDAG,AI,51,8,0,0
7,Gemma 2,HLTC,AI,42,17,0,0
8,Tinyllama,LDAG,human,11,4,34,10
9,Tinyllama,HLTC,human,13,5,30,11
10,Tinyllama,LDAG,AI,11,5,31,12
11,Tinyllama,HLTC,AI,11,3,35,10
1 model name dataname expected human AI unsure invalid
2 0 Microsoft Phi LDAG human 13 14 12 20
3 1 Microsoft Phi HLTC human 12 11 19 17
4 2 Microsoft Phi LDAG AI 10 18 14 17
5 3 Microsoft Phi HLTC AI 14 13 23 9
6 4 Gemma 2 LDAG human 35 24 0 0
7 5 Gemma 2 HLTC human 45 14 0 0
8 6 Gemma 2 LDAG AI 51 8 0 0
9 7 Gemma 2 HLTC AI 42 17 0 0
10 8 Tinyllama LDAG human 11 4 34 10
11 9 Tinyllama HLTC human 13 5 30 11
12 10 Tinyllama LDAG AI 11 5 31 12
13 11 Tinyllama HLTC AI 11 3 35 10

13
tests/outputs/results (Multi-Shot).CSV Normal file
View file

@ -0,0 +1,13 @@
,model name,dataname,expected,human,AI,unsure,invalid
0,Microsoft Phi,LDAG,human,1,10,46,2
1,Microsoft Phi,HLTC,human,7,6,44,2
2,Microsoft Phi,LDAG,AI,7,4,43,5
3,Microsoft Phi,HLTC,AI,4,7,47,1
4,Gemma 2,LDAG,human,24,35,0,0
5,Gemma 2,HLTC,human,48,11,0,0
6,Gemma 2,LDAG,AI,49,10,0,0
7,Gemma 2,HLTC,AI,48,11,0,0
8,Tinyllama,LDAG,human,24,5,28,2
9,Tinyllama,HLTC,human,17,3,36,3
10,Tinyllama,LDAG,AI,14,3,37,5
11,Tinyllama,HLTC,AI,17,1,39,2
1 model name dataname expected human AI unsure invalid
2 0 Microsoft Phi LDAG human 1 10 46 2
3 1 Microsoft Phi HLTC human 7 6 44 2
4 2 Microsoft Phi LDAG AI 7 4 43 5
5 3 Microsoft Phi HLTC AI 4 7 47 1
6 4 Gemma 2 LDAG human 24 35 0 0
7 5 Gemma 2 HLTC human 48 11 0 0
8 6 Gemma 2 LDAG AI 49 10 0 0
9 7 Gemma 2 HLTC AI 48 11 0 0
10 8 Tinyllama LDAG human 24 5 28 2
11 9 Tinyllama HLTC human 17 3 36 3
12 10 Tinyllama LDAG AI 14 3 37 5
13 11 Tinyllama HLTC AI 17 1 39 2

13
tests/outputs/results (classic).CSV Normal file
View file

@ -0,0 +1,13 @@
,model name,dataname,expected,human,AI,unsure,invalid
0,Microsoft Phi,LDAG,human,3,4,48,4
1,Microsoft Phi,HLTC,human,7,6,42,4
2,Microsoft Phi,LDAG,AI,5,9,43,2
3,Microsoft Phi,HLTC,AI,4,4,49,2
4,Gemma 2,LDAG,human,38,21,0,0
5,Gemma 2,HLTC,human,47,12,0,0
6,Gemma 2,LDAG,AI,51,8,0,0
7,Gemma 2,HLTC,AI,46,13,0,0
8,Tinyllama,LDAG,human,13,3,37,6
9,Tinyllama,HLTC,human,9,3,40,7
10,Tinyllama,LDAG,AI,8,5,40,6
11,Tinyllama,HLTC,AI,11,4,37,7
1 model name dataname expected human AI unsure invalid
2 0 Microsoft Phi LDAG human 3 4 48 4
3 1 Microsoft Phi HLTC human 7 6 42 4
4 2 Microsoft Phi LDAG AI 5 9 43 2
5 3 Microsoft Phi HLTC AI 4 4 49 2
6 4 Gemma 2 LDAG human 38 21 0 0
7 5 Gemma 2 HLTC human 47 12 0 0
8 6 Gemma 2 LDAG AI 51 8 0 0
9 7 Gemma 2 HLTC AI 46 13 0 0
10 8 Tinyllama LDAG human 13 3 37 6
11 9 Tinyllama HLTC human 9 3 40 7
12 10 Tinyllama LDAG AI 8 5 40 6
13 11 Tinyllama HLTC AI 11 4 37 7