Muennighoff commited on
Commit
79391b8
1 Parent(s): fdef977
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0.csv +21 -0
  2. evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0.json +87 -0
  3. evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0_lm-eval_global_step52452_2023-02-09-23-24-23_0shots_backup.json +87 -0
  4. evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_1.json +87 -0
  5. evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_1_lm-eval_global_step52452_2023-02-09-23-24-23_1shots_backup.json +87 -0
  6. evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_2.json +87 -0
  7. evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_2_lm-eval_global_step52452_2023-02-09-23-24-23_2shots_backup.json +87 -0
  8. evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_3.json +87 -0
  9. evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_3_lm-eval_global_step52452_2023-02-09-23-24-23_3shots_backup.json +87 -0
  10. evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_4.json +87 -0
  11. evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_4_lm-eval_global_step52452_2023-02-09-23-24-23_4shots_backup.json +87 -0
  12. evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_5.json +87 -0
  13. evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_5_lm-eval_global_step52452_2023-02-09-23-24-23_5shots_backup.json +87 -0
  14. evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0.csv +21 -0
  15. evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0.json +87 -0
  16. evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0_lm-eval_global_step52452_2023-02-09-17-38-12_0shots_backup.json +87 -0
  17. evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_1.json +87 -0
  18. evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_1_lm-eval_global_step52452_2023-02-09-17-38-12_1shots_backup.json +87 -0
  19. evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_2.json +87 -0
  20. evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_2_lm-eval_global_step52452_2023-02-09-17-36-57_2shots_backup.json +87 -0
  21. evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_3.json +87 -0
  22. evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_3_lm-eval_global_step52452_2023-02-09-17-36-57_3shots_backup.json +87 -0
  23. evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_4.json +87 -0
  24. evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_4_lm-eval_global_step52452_2023-02-09-17-38-12_4shots_backup.json +87 -0
  25. evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_5.json +87 -0
  26. evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_5_lm-eval_global_step52452_2023-02-09-17-38-12_5shots_backup.json +87 -0
  27. evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_0.csv +21 -0
  28. evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_0.json +87 -0
  29. evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_0_lm-eval_global_step42000_2023-02-08-13-42-29_0shots_backup.json +87 -0
  30. evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_1.json +87 -0
  31. evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_1_lm-eval_global_step42000_2023-02-08-13-42-29_1shots_backup.json +87 -0
  32. evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_2.json +87 -0
  33. evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_2_lm-eval_global_step42000_2023-02-08-13-42-29_2shots_backup.json +87 -0
  34. evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_3.json +87 -0
  35. evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_3_lm-eval_global_step42000_2023-02-08-13-42-29_3shots_backup.json +87 -0
  36. evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_4.json +87 -0
  37. evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_4_lm-eval_global_step42000_2023-02-08-13-42-29_4shots_backup.json +87 -0
  38. evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_5.json +87 -0
  39. evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_5_lm-eval_global_step42000_2023-02-08-13-42-29_5shots_backup.json +87 -0
  40. evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0.csv +21 -0
  41. evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0.json +87 -0
  42. evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0_lm-eval_global_step52452_2023-02-09-23-08-31_0shots_backup.json +87 -0
  43. evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_1.json +87 -0
  44. evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_1_lm-eval_global_step52452_2023-02-09-23-08-31_1shots_backup.json +87 -0
  45. evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_2.json +87 -0
  46. evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_2_lm-eval_global_step52452_2023-02-09-23-08-31_2shots_backup.json +87 -0
  47. evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_3.json +87 -0
  48. evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_3_lm-eval_global_step52452_2023-02-09-23-08-31_3shots_backup.json +87 -0
  49. evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_4.json +87 -0
  50. evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_4_lm-eval_global_step52452_2023-02-09-23-08-31_4shots_backup.json +87 -0
evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0.csv ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ task,metric,value,err,version
2
+ anli_r1,acc,0.339,0.01497675877162034,0
3
+ anli_r2,acc,0.335,0.014933117490932573,0
4
+ anli_r3,acc,0.33416666666666667,0.013622434813136783,0
5
+ arc_challenge,acc,0.17918088737201365,0.011207045216615674,0
6
+ arc_challenge,acc_norm,0.2235494880546075,0.012174896631202614,0
7
+ arc_easy,acc,0.4335016835016835,0.010168640625454107,0
8
+ arc_easy,acc_norm,0.3846801346801347,0.009983171707009006,0
9
+ boolq,acc,0.5938837920489297,0.008589510943787407,1
10
+ cb,acc,0.4107142857142857,0.0663363415035954,1
11
+ cb,f1,0.1940928270042194,,1
12
+ copa,acc,0.62,0.04878317312145632,0
13
+ hellaswag,acc,0.2951603266281617,0.004551826272978059,0
14
+ hellaswag,acc_norm,0.3241386178052181,0.004670955399641126,0
15
+ piqa,acc,0.6218715995647442,0.011313980666854535,0
16
+ piqa,acc_norm,0.6267682263329706,0.011284653078254898,0
17
+ rte,acc,0.5234657039711191,0.03006330041190266,0
18
+ sciq,acc,0.732,0.01401329270272948,0
19
+ sciq,acc_norm,0.669,0.01488827258820394,0
20
+ storycloze_2016,acc,0.5873864243719936,0.011384472322969045,0
21
+ winogrande,acc,0.5059194948697711,0.01405150083848581,0
evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.339,
5
+ "acc_stderr": 0.01497675877162034
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.335,
9
+ "acc_stderr": 0.014933117490932573
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.33416666666666667,
13
+ "acc_stderr": 0.013622434813136783
14
+ },
15
+ "cb": {
16
+ "acc": 0.4107142857142857,
17
+ "acc_stderr": 0.0663363415035954,
18
+ "f1": 0.1940928270042194
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145632
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2951603266281617,
26
+ "acc_stderr": 0.004551826272978059,
27
+ "acc_norm": 0.3241386178052181,
28
+ "acc_norm_stderr": 0.004670955399641126
29
+ },
30
+ "rte": {
31
+ "acc": 0.5234657039711191,
32
+ "acc_stderr": 0.03006330041190266
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5059194948697711,
36
+ "acc_stderr": 0.01405150083848581
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5873864243719936,
40
+ "acc_stderr": 0.011384472322969045
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5938837920489297,
44
+ "acc_stderr": 0.008589510943787407
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4335016835016835,
48
+ "acc_stderr": 0.010168640625454107,
49
+ "acc_norm": 0.3846801346801347,
50
+ "acc_norm_stderr": 0.009983171707009006
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.17918088737201365,
54
+ "acc_stderr": 0.011207045216615674,
55
+ "acc_norm": 0.2235494880546075,
56
+ "acc_norm_stderr": 0.012174896631202614
57
+ },
58
+ "sciq": {
59
+ "acc": 0.732,
60
+ "acc_stderr": 0.01401329270272948,
61
+ "acc_norm": 0.669,
62
+ "acc_norm_stderr": 0.01488827258820394
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6218715995647442,
66
+ "acc_stderr": 0.011313980666854535,
67
+ "acc_norm": 0.6267682263329706,
68
+ "acc_norm_stderr": 0.011284653078254898
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0_lm-eval_global_step52452_2023-02-09-23-24-23_0shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.339,
5
+ "acc_stderr": 0.01497675877162034
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.335,
9
+ "acc_stderr": 0.014933117490932573
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.33416666666666667,
13
+ "acc_stderr": 0.013622434813136783
14
+ },
15
+ "cb": {
16
+ "acc": 0.4107142857142857,
17
+ "acc_stderr": 0.0663363415035954,
18
+ "f1": 0.1940928270042194
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145632
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2951603266281617,
26
+ "acc_stderr": 0.004551826272978059,
27
+ "acc_norm": 0.3241386178052181,
28
+ "acc_norm_stderr": 0.004670955399641126
29
+ },
30
+ "rte": {
31
+ "acc": 0.5234657039711191,
32
+ "acc_stderr": 0.03006330041190266
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5059194948697711,
36
+ "acc_stderr": 0.01405150083848581
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5873864243719936,
40
+ "acc_stderr": 0.011384472322969045
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5938837920489297,
44
+ "acc_stderr": 0.008589510943787407
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4335016835016835,
48
+ "acc_stderr": 0.010168640625454107,
49
+ "acc_norm": 0.3846801346801347,
50
+ "acc_norm_stderr": 0.009983171707009006
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.17918088737201365,
54
+ "acc_stderr": 0.011207045216615674,
55
+ "acc_norm": 0.2235494880546075,
56
+ "acc_norm_stderr": 0.012174896631202614
57
+ },
58
+ "sciq": {
59
+ "acc": 0.732,
60
+ "acc_stderr": 0.01401329270272948,
61
+ "acc_norm": 0.669,
62
+ "acc_norm_stderr": 0.01488827258820394
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6218715995647442,
66
+ "acc_stderr": 0.011313980666854535,
67
+ "acc_norm": 0.6267682263329706,
68
+ "acc_norm_stderr": 0.011284653078254898
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_1.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.36,
5
+ "acc_stderr": 0.015186527932040122
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.349,
9
+ "acc_stderr": 0.015080663991563102
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.32166666666666666,
13
+ "acc_stderr": 0.01349009528298952
14
+ },
15
+ "cb": {
16
+ "acc": 0.4107142857142857,
17
+ "acc_stderr": 0.0663363415035954,
18
+ "f1": 0.28651292802236195
19
+ },
20
+ "copa": {
21
+ "acc": 0.64,
22
+ "acc_stderr": 0.048241815132442176
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2913762198765186,
26
+ "acc_stderr": 0.004534677750102734,
27
+ "acc_norm": 0.3249352718581956,
28
+ "acc_norm_stderr": 0.0046739348371504464
29
+ },
30
+ "rte": {
31
+ "acc": 0.5523465703971119,
32
+ "acc_stderr": 0.02993107036293953
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.494869771112865,
36
+ "acc_stderr": 0.014051745961790513
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5793693212185996,
40
+ "acc_stderr": 0.01141582799434265
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5110091743119266,
44
+ "acc_stderr": 0.008742934884517647
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4297138047138047,
48
+ "acc_stderr": 0.010157908005763678,
49
+ "acc_norm": 0.3792087542087542,
50
+ "acc_norm_stderr": 0.00995589166886556
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.1825938566552901,
54
+ "acc_stderr": 0.011289730684564982,
55
+ "acc_norm": 0.21928327645051193,
56
+ "acc_norm_stderr": 0.012091245787615734
57
+ },
58
+ "sciq": {
59
+ "acc": 0.705,
60
+ "acc_stderr": 0.014428554438445517,
61
+ "acc_norm": 0.658,
62
+ "acc_norm_stderr": 0.015008706182121731
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6218715995647442,
66
+ "acc_stderr": 0.011313980666854535,
67
+ "acc_norm": 0.6109902067464635,
68
+ "acc_norm_stderr": 0.011374774974447464
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_1_lm-eval_global_step52452_2023-02-09-23-24-23_1shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.36,
5
+ "acc_stderr": 0.015186527932040122
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.349,
9
+ "acc_stderr": 0.015080663991563102
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.32166666666666666,
13
+ "acc_stderr": 0.01349009528298952
14
+ },
15
+ "cb": {
16
+ "acc": 0.4107142857142857,
17
+ "acc_stderr": 0.0663363415035954,
18
+ "f1": 0.28651292802236195
19
+ },
20
+ "copa": {
21
+ "acc": 0.64,
22
+ "acc_stderr": 0.048241815132442176
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2913762198765186,
26
+ "acc_stderr": 0.004534677750102734,
27
+ "acc_norm": 0.3249352718581956,
28
+ "acc_norm_stderr": 0.0046739348371504464
29
+ },
30
+ "rte": {
31
+ "acc": 0.5523465703971119,
32
+ "acc_stderr": 0.02993107036293953
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.494869771112865,
36
+ "acc_stderr": 0.014051745961790513
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5793693212185996,
40
+ "acc_stderr": 0.01141582799434265
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5110091743119266,
44
+ "acc_stderr": 0.008742934884517647
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4297138047138047,
48
+ "acc_stderr": 0.010157908005763678,
49
+ "acc_norm": 0.3792087542087542,
50
+ "acc_norm_stderr": 0.00995589166886556
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.1825938566552901,
54
+ "acc_stderr": 0.011289730684564982,
55
+ "acc_norm": 0.21928327645051193,
56
+ "acc_norm_stderr": 0.012091245787615734
57
+ },
58
+ "sciq": {
59
+ "acc": 0.705,
60
+ "acc_stderr": 0.014428554438445517,
61
+ "acc_norm": 0.658,
62
+ "acc_norm_stderr": 0.015008706182121731
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6218715995647442,
66
+ "acc_stderr": 0.011313980666854535,
67
+ "acc_norm": 0.6109902067464635,
68
+ "acc_norm_stderr": 0.011374774974447464
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_2.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.311,
5
+ "acc_stderr": 0.014645596385722695
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.356,
9
+ "acc_stderr": 0.015149042659306625
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.33666666666666667,
13
+ "acc_stderr": 0.01364760294240639
14
+ },
15
+ "cb": {
16
+ "acc": 0.4642857142857143,
17
+ "acc_stderr": 0.06724777654937658,
18
+ "f1": 0.316548463356974
19
+ },
20
+ "copa": {
21
+ "acc": 0.63,
22
+ "acc_stderr": 0.048523658709391
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29047998406691894,
26
+ "acc_stderr": 0.004530560646902538,
27
+ "acc_norm": 0.3179645488946425,
28
+ "acc_norm_stderr": 0.004647338877642189
29
+ },
30
+ "rte": {
31
+ "acc": 0.48736462093862815,
32
+ "acc_stderr": 0.030086851767188564
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5098658247829518,
36
+ "acc_stderr": 0.014049749833367596
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5793693212185996,
40
+ "acc_stderr": 0.011415827994342655
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4746177370030581,
44
+ "acc_stderr": 0.008733779541853504
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42297979797979796,
48
+ "acc_stderr": 0.010137328382209104,
49
+ "acc_norm": 0.39057239057239057,
50
+ "acc_norm_stderr": 0.010011059112064229
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.18515358361774745,
54
+ "acc_stderr": 0.011350774438389699,
55
+ "acc_norm": 0.22525597269624573,
56
+ "acc_norm_stderr": 0.01220783999540731
57
+ },
58
+ "sciq": {
59
+ "acc": 0.727,
60
+ "acc_stderr": 0.014095022868717607,
61
+ "acc_norm": 0.677,
62
+ "acc_norm_stderr": 0.014794927843348635
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6316648531011969,
66
+ "acc_stderr": 0.011254089354334373,
67
+ "acc_norm": 0.6294885745375408,
68
+ "acc_norm_stderr": 0.01126782647544766
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_2_lm-eval_global_step52452_2023-02-09-23-24-23_2shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.311,
5
+ "acc_stderr": 0.014645596385722695
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.356,
9
+ "acc_stderr": 0.015149042659306625
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.33666666666666667,
13
+ "acc_stderr": 0.01364760294240639
14
+ },
15
+ "cb": {
16
+ "acc": 0.4642857142857143,
17
+ "acc_stderr": 0.06724777654937658,
18
+ "f1": 0.316548463356974
19
+ },
20
+ "copa": {
21
+ "acc": 0.63,
22
+ "acc_stderr": 0.048523658709391
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29047998406691894,
26
+ "acc_stderr": 0.004530560646902538,
27
+ "acc_norm": 0.3179645488946425,
28
+ "acc_norm_stderr": 0.004647338877642189
29
+ },
30
+ "rte": {
31
+ "acc": 0.48736462093862815,
32
+ "acc_stderr": 0.030086851767188564
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5098658247829518,
36
+ "acc_stderr": 0.014049749833367596
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5793693212185996,
40
+ "acc_stderr": 0.011415827994342655
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4746177370030581,
44
+ "acc_stderr": 0.008733779541853504
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42297979797979796,
48
+ "acc_stderr": 0.010137328382209104,
49
+ "acc_norm": 0.39057239057239057,
50
+ "acc_norm_stderr": 0.010011059112064229
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.18515358361774745,
54
+ "acc_stderr": 0.011350774438389699,
55
+ "acc_norm": 0.22525597269624573,
56
+ "acc_norm_stderr": 0.01220783999540731
57
+ },
58
+ "sciq": {
59
+ "acc": 0.727,
60
+ "acc_stderr": 0.014095022868717607,
61
+ "acc_norm": 0.677,
62
+ "acc_norm_stderr": 0.014794927843348635
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6316648531011969,
66
+ "acc_stderr": 0.011254089354334373,
67
+ "acc_norm": 0.6294885745375408,
68
+ "acc_norm_stderr": 0.01126782647544766
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_3.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.312,
5
+ "acc_stderr": 0.014658474370509008
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.344,
9
+ "acc_stderr": 0.015029633724408947
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3425,
13
+ "acc_stderr": 0.013704669762934725
14
+ },
15
+ "cb": {
16
+ "acc": 0.5178571428571429,
17
+ "acc_stderr": 0.06737697508644647,
18
+ "f1": 0.3422885572139303
19
+ },
20
+ "copa": {
21
+ "acc": 0.6,
22
+ "acc_stderr": 0.049236596391733084
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.291575383389763,
26
+ "acc_stderr": 0.004535589759202657,
27
+ "acc_norm": 0.32284405496912966,
28
+ "acc_norm_stderr": 0.004666080865179641
29
+ },
30
+ "rte": {
31
+ "acc": 0.5054151624548736,
32
+ "acc_stderr": 0.030094698123239966
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5011838989739542,
36
+ "acc_stderr": 0.014052446290529015
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5740245857830037,
40
+ "acc_stderr": 0.011435014262181197
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4688073394495413,
44
+ "acc_stderr": 0.008728020822889253
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42424242424242425,
48
+ "acc_stderr": 0.010141333654958574,
49
+ "acc_norm": 0.38425925925925924,
50
+ "acc_norm_stderr": 0.009981120724601443
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.18686006825938567,
54
+ "acc_stderr": 0.011391015649694391,
55
+ "acc_norm": 0.22440273037542663,
56
+ "acc_norm_stderr": 0.012191404938603838
57
+ },
58
+ "sciq": {
59
+ "acc": 0.723,
60
+ "acc_stderr": 0.014158794845306265,
61
+ "acc_norm": 0.682,
62
+ "acc_norm_stderr": 0.014734079309311901
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6284004352557128,
66
+ "acc_stderr": 0.011274603006724743,
67
+ "acc_norm": 0.6196953210010882,
68
+ "acc_norm_stderr": 0.011326620892570314
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_3_lm-eval_global_step52452_2023-02-09-23-24-23_3shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.312,
5
+ "acc_stderr": 0.014658474370509008
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.344,
9
+ "acc_stderr": 0.015029633724408947
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3425,
13
+ "acc_stderr": 0.013704669762934725
14
+ },
15
+ "cb": {
16
+ "acc": 0.5178571428571429,
17
+ "acc_stderr": 0.06737697508644647,
18
+ "f1": 0.3422885572139303
19
+ },
20
+ "copa": {
21
+ "acc": 0.6,
22
+ "acc_stderr": 0.049236596391733084
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.291575383389763,
26
+ "acc_stderr": 0.004535589759202657,
27
+ "acc_norm": 0.32284405496912966,
28
+ "acc_norm_stderr": 0.004666080865179641
29
+ },
30
+ "rte": {
31
+ "acc": 0.5054151624548736,
32
+ "acc_stderr": 0.030094698123239966
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5011838989739542,
36
+ "acc_stderr": 0.014052446290529015
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5740245857830037,
40
+ "acc_stderr": 0.011435014262181197
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4688073394495413,
44
+ "acc_stderr": 0.008728020822889253
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42424242424242425,
48
+ "acc_stderr": 0.010141333654958574,
49
+ "acc_norm": 0.38425925925925924,
50
+ "acc_norm_stderr": 0.009981120724601443
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.18686006825938567,
54
+ "acc_stderr": 0.011391015649694391,
55
+ "acc_norm": 0.22440273037542663,
56
+ "acc_norm_stderr": 0.012191404938603838
57
+ },
58
+ "sciq": {
59
+ "acc": 0.723,
60
+ "acc_stderr": 0.014158794845306265,
61
+ "acc_norm": 0.682,
62
+ "acc_norm_stderr": 0.014734079309311901
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6284004352557128,
66
+ "acc_stderr": 0.011274603006724743,
67
+ "acc_norm": 0.6196953210010882,
68
+ "acc_norm_stderr": 0.011326620892570314
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_4.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.336,
5
+ "acc_stderr": 0.014944140233795025
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.352,
9
+ "acc_stderr": 0.01511040450564867
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3383333333333333,
13
+ "acc_stderr": 0.013664144006618266
14
+ },
15
+ "cb": {
16
+ "acc": 0.5178571428571429,
17
+ "acc_stderr": 0.06737697508644645,
18
+ "f1": 0.3362023995826813
19
+ },
20
+ "copa": {
21
+ "acc": 0.6,
22
+ "acc_stderr": 0.049236596391733084
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29645488946425014,
26
+ "acc_stderr": 0.004557606227194286,
27
+ "acc_norm": 0.3234415455088628,
28
+ "acc_norm_stderr": 0.004668335725410298
29
+ },
30
+ "rte": {
31
+ "acc": 0.49458483754512633,
32
+ "acc_stderr": 0.030094698123239966
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.4877663772691397,
36
+ "acc_stderr": 0.01404827882040562
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5788348476750401,
40
+ "acc_stderr": 0.011417808278216117
41
+ },
42
+ "boolq": {
43
+ "acc": 0.44434250764525995,
44
+ "acc_stderr": 0.00869070599067338
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42845117845117847,
48
+ "acc_stderr": 0.010154195733990975,
49
+ "acc_norm": 0.3930976430976431,
50
+ "acc_norm_stderr": 0.010022540618945312
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.1962457337883959,
54
+ "acc_stderr": 0.01160601988141629,
55
+ "acc_norm": 0.22781569965870307,
56
+ "acc_norm_stderr": 0.012256708602326905
57
+ },
58
+ "sciq": {
59
+ "acc": 0.719,
60
+ "acc_stderr": 0.014221154708434929,
61
+ "acc_norm": 0.686,
62
+ "acc_norm_stderr": 0.014683991951087967
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6273122959738846,
66
+ "acc_stderr": 0.01128131833289774,
67
+ "acc_norm": 0.6169749727965179,
68
+ "acc_norm_stderr": 0.01134208170908285
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_4_lm-eval_global_step52452_2023-02-09-23-24-23_4shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.336,
5
+ "acc_stderr": 0.014944140233795025
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.352,
9
+ "acc_stderr": 0.01511040450564867
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3383333333333333,
13
+ "acc_stderr": 0.013664144006618266
14
+ },
15
+ "cb": {
16
+ "acc": 0.5178571428571429,
17
+ "acc_stderr": 0.06737697508644645,
18
+ "f1": 0.3362023995826813
19
+ },
20
+ "copa": {
21
+ "acc": 0.6,
22
+ "acc_stderr": 0.049236596391733084
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29645488946425014,
26
+ "acc_stderr": 0.004557606227194286,
27
+ "acc_norm": 0.3234415455088628,
28
+ "acc_norm_stderr": 0.004668335725410298
29
+ },
30
+ "rte": {
31
+ "acc": 0.49458483754512633,
32
+ "acc_stderr": 0.030094698123239966
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.4877663772691397,
36
+ "acc_stderr": 0.01404827882040562
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5788348476750401,
40
+ "acc_stderr": 0.011417808278216117
41
+ },
42
+ "boolq": {
43
+ "acc": 0.44434250764525995,
44
+ "acc_stderr": 0.00869070599067338
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42845117845117847,
48
+ "acc_stderr": 0.010154195733990975,
49
+ "acc_norm": 0.3930976430976431,
50
+ "acc_norm_stderr": 0.010022540618945312
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.1962457337883959,
54
+ "acc_stderr": 0.01160601988141629,
55
+ "acc_norm": 0.22781569965870307,
56
+ "acc_norm_stderr": 0.012256708602326905
57
+ },
58
+ "sciq": {
59
+ "acc": 0.719,
60
+ "acc_stderr": 0.014221154708434929,
61
+ "acc_norm": 0.686,
62
+ "acc_norm_stderr": 0.014683991951087967
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6273122959738846,
66
+ "acc_stderr": 0.01128131833289774,
67
+ "acc_norm": 0.6169749727965179,
68
+ "acc_norm_stderr": 0.01134208170908285
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_5.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.375,
5
+ "acc_stderr": 0.015316971293620996
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.339,
9
+ "acc_stderr": 0.014976758771620344
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3408333333333333,
13
+ "acc_stderr": 0.01368860079329693
14
+ },
15
+ "cb": {
16
+ "acc": 0.5535714285714286,
17
+ "acc_stderr": 0.06703189227942395,
18
+ "f1": 0.3464373464373464
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145633
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29486158135829516,
26
+ "acc_stderr": 0.0045504861860190746,
27
+ "acc_norm": 0.32304321848237405,
28
+ "acc_norm_stderr": 0.0046668334527961925
29
+ },
30
+ "rte": {
31
+ "acc": 0.5018050541516246,
32
+ "acc_stderr": 0.030096267148976626
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5027624309392266,
36
+ "acc_stderr": 0.014052271211616441
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5809727418492785,
40
+ "acc_stderr": 0.011409804749706194
41
+ },
42
+ "boolq": {
43
+ "acc": 0.44128440366972477,
44
+ "acc_stderr": 0.008684548127832634
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4297138047138047,
48
+ "acc_stderr": 0.010157908005763676,
49
+ "acc_norm": 0.3985690235690236,
50
+ "acc_norm_stderr": 0.010046455400477931
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19197952218430034,
54
+ "acc_stderr": 0.011509598906598112,
55
+ "acc_norm": 0.22525597269624573,
56
+ "acc_norm_stderr": 0.012207839995407303
57
+ },
58
+ "sciq": {
59
+ "acc": 0.721,
60
+ "acc_stderr": 0.014190150117612032,
61
+ "acc_norm": 0.682,
62
+ "acc_norm_stderr": 0.014734079309311901
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6305767138193689,
66
+ "acc_stderr": 0.011260988628572347,
67
+ "acc_norm": 0.6180631120783461,
68
+ "acc_norm_stderr": 0.011335942557505228
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_r_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_5_lm-eval_global_step52452_2023-02-09-23-24-23_5shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.375,
5
+ "acc_stderr": 0.015316971293620996
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.339,
9
+ "acc_stderr": 0.014976758771620344
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3408333333333333,
13
+ "acc_stderr": 0.01368860079329693
14
+ },
15
+ "cb": {
16
+ "acc": 0.5535714285714286,
17
+ "acc_stderr": 0.06703189227942395,
18
+ "f1": 0.3464373464373464
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145633
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29486158135829516,
26
+ "acc_stderr": 0.0045504861860190746,
27
+ "acc_norm": 0.32304321848237405,
28
+ "acc_norm_stderr": 0.0046668334527961925
29
+ },
30
+ "rte": {
31
+ "acc": 0.5018050541516246,
32
+ "acc_stderr": 0.030096267148976626
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5027624309392266,
36
+ "acc_stderr": 0.014052271211616441
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5809727418492785,
40
+ "acc_stderr": 0.011409804749706194
41
+ },
42
+ "boolq": {
43
+ "acc": 0.44128440366972477,
44
+ "acc_stderr": 0.008684548127832634
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4297138047138047,
48
+ "acc_stderr": 0.010157908005763676,
49
+ "acc_norm": 0.3985690235690236,
50
+ "acc_norm_stderr": 0.010046455400477931
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19197952218430034,
54
+ "acc_stderr": 0.011509598906598112,
55
+ "acc_norm": 0.22525597269624573,
56
+ "acc_norm_stderr": 0.012207839995407303
57
+ },
58
+ "sciq": {
59
+ "acc": 0.721,
60
+ "acc_stderr": 0.014190150117612032,
61
+ "acc_norm": 0.682,
62
+ "acc_norm_stderr": 0.014734079309311901
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6305767138193689,
66
+ "acc_stderr": 0.011260988628572347,
67
+ "acc_norm": 0.6180631120783461,
68
+ "acc_norm_stderr": 0.011335942557505228
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0.csv ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ task,metric,value,err,version
2
+ anli_r1,acc,0.339,0.01497675877162034,0
3
+ anli_r2,acc,0.336,0.014944140233795027,0
4
+ anli_r3,acc,0.3358333333333333,0.01363926119093288,0
5
+ arc_challenge,acc,0.1885665529010239,0.011430897647675803,0
6
+ arc_challenge,acc_norm,0.22610921501706485,0.01222420209706328,0
7
+ arc_easy,acc,0.43308080808080807,0.010167478013701799,0
8
+ arc_easy,acc_norm,0.38173400673400676,0.009968648851839667,0
9
+ boolq,acc,0.5944954128440367,0.008587459055441612,1
10
+ cb,acc,0.4107142857142857,0.0663363415035954,1
11
+ cb,f1,0.1940928270042194,,1
12
+ copa,acc,0.63,0.04852365870939099,0
13
+ hellaswag,acc,0.297450707030472,0.004562022467161891,0
14
+ hellaswag,acc_norm,0.32374029077872934,0.004669459891917689,0
15
+ piqa,acc,0.6158868335146899,0.011348160741479148,0
16
+ piqa,acc_norm,0.6218715995647442,0.011313980666854533,0
17
+ rte,acc,0.5234657039711191,0.03006330041190266,0
18
+ sciq,acc,0.735,0.013963164754809953,0
19
+ sciq,acc_norm,0.668,0.014899597242811476,0
20
+ storycloze_2016,acc,0.5905932656333511,0.01137105952719707,0
21
+ winogrande,acc,0.5090765588003157,0.014050170094497707,0
evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.339,
5
+ "acc_stderr": 0.01497675877162034
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.336,
9
+ "acc_stderr": 0.014944140233795027
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3358333333333333,
13
+ "acc_stderr": 0.01363926119093288
14
+ },
15
+ "cb": {
16
+ "acc": 0.4107142857142857,
17
+ "acc_stderr": 0.0663363415035954,
18
+ "f1": 0.1940928270042194
19
+ },
20
+ "copa": {
21
+ "acc": 0.63,
22
+ "acc_stderr": 0.04852365870939099
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.297450707030472,
26
+ "acc_stderr": 0.004562022467161891,
27
+ "acc_norm": 0.32374029077872934,
28
+ "acc_norm_stderr": 0.004669459891917689
29
+ },
30
+ "rte": {
31
+ "acc": 0.5234657039711191,
32
+ "acc_stderr": 0.03006330041190266
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5090765588003157,
36
+ "acc_stderr": 0.014050170094497707
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5905932656333511,
40
+ "acc_stderr": 0.01137105952719707
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5944954128440367,
44
+ "acc_stderr": 0.008587459055441612
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.43308080808080807,
48
+ "acc_stderr": 0.010167478013701799,
49
+ "acc_norm": 0.38173400673400676,
50
+ "acc_norm_stderr": 0.009968648851839667
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.1885665529010239,
54
+ "acc_stderr": 0.011430897647675803,
55
+ "acc_norm": 0.22610921501706485,
56
+ "acc_norm_stderr": 0.01222420209706328
57
+ },
58
+ "sciq": {
59
+ "acc": 0.735,
60
+ "acc_stderr": 0.013963164754809953,
61
+ "acc_norm": 0.668,
62
+ "acc_norm_stderr": 0.014899597242811476
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6158868335146899,
66
+ "acc_stderr": 0.011348160741479148,
67
+ "acc_norm": 0.6218715995647442,
68
+ "acc_norm_stderr": 0.011313980666854533
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0_lm-eval_global_step52452_2023-02-09-17-38-12_0shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.339,
5
+ "acc_stderr": 0.01497675877162034
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.336,
9
+ "acc_stderr": 0.014944140233795027
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3358333333333333,
13
+ "acc_stderr": 0.01363926119093288
14
+ },
15
+ "cb": {
16
+ "acc": 0.4107142857142857,
17
+ "acc_stderr": 0.0663363415035954,
18
+ "f1": 0.1940928270042194
19
+ },
20
+ "copa": {
21
+ "acc": 0.63,
22
+ "acc_stderr": 0.04852365870939099
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.297450707030472,
26
+ "acc_stderr": 0.004562022467161891,
27
+ "acc_norm": 0.32374029077872934,
28
+ "acc_norm_stderr": 0.004669459891917689
29
+ },
30
+ "rte": {
31
+ "acc": 0.5234657039711191,
32
+ "acc_stderr": 0.03006330041190266
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5090765588003157,
36
+ "acc_stderr": 0.014050170094497707
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5905932656333511,
40
+ "acc_stderr": 0.01137105952719707
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5944954128440367,
44
+ "acc_stderr": 0.008587459055441612
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.43308080808080807,
48
+ "acc_stderr": 0.010167478013701799,
49
+ "acc_norm": 0.38173400673400676,
50
+ "acc_norm_stderr": 0.009968648851839667
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.1885665529010239,
54
+ "acc_stderr": 0.011430897647675803,
55
+ "acc_norm": 0.22610921501706485,
56
+ "acc_norm_stderr": 0.01222420209706328
57
+ },
58
+ "sciq": {
59
+ "acc": 0.735,
60
+ "acc_stderr": 0.013963164754809953,
61
+ "acc_norm": 0.668,
62
+ "acc_norm_stderr": 0.014899597242811476
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6158868335146899,
66
+ "acc_stderr": 0.011348160741479148,
67
+ "acc_norm": 0.6218715995647442,
68
+ "acc_norm_stderr": 0.011313980666854533
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_1.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.358,
5
+ "acc_stderr": 0.015167928865407559
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.35,
9
+ "acc_stderr": 0.015090650341444236
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.32,
13
+ "acc_stderr": 0.013471620929769152
14
+ },
15
+ "cb": {
16
+ "acc": 0.4107142857142857,
17
+ "acc_stderr": 0.0663363415035954,
18
+ "f1": 0.28651292802236195
19
+ },
20
+ "copa": {
21
+ "acc": 0.68,
22
+ "acc_stderr": 0.04688261722621504
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2930691097390958,
26
+ "acc_stderr": 0.004542396269999213,
27
+ "acc_norm": 0.3207528380800637,
28
+ "acc_norm_stderr": 0.004658120152230808
29
+ },
30
+ "rte": {
31
+ "acc": 0.5631768953068592,
32
+ "acc_stderr": 0.02985524739031495
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.4964483030781373,
36
+ "acc_stderr": 0.01405213114691586
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5793693212185996,
40
+ "acc_stderr": 0.01141582799434265
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5125382262996941,
44
+ "acc_stderr": 0.008742304974218311
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4313973063973064,
48
+ "acc_stderr": 0.010162752847747498,
49
+ "acc_norm": 0.38341750841750843,
50
+ "acc_norm_stderr": 0.009976995068264717
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19027303754266212,
54
+ "acc_stderr": 0.011470424179225709,
55
+ "acc_norm": 0.22610921501706485,
56
+ "acc_norm_stderr": 0.012224202097063274
57
+ },
58
+ "sciq": {
59
+ "acc": 0.704,
60
+ "acc_stderr": 0.014442734941575022,
61
+ "acc_norm": 0.658,
62
+ "acc_norm_stderr": 0.015008706182121731
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6240478781284005,
66
+ "acc_stderr": 0.011301098166895732,
67
+ "acc_norm": 0.6158868335146899,
68
+ "acc_norm_stderr": 0.011348160741479136
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_1_lm-eval_global_step52452_2023-02-09-17-38-12_1shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.358,
5
+ "acc_stderr": 0.015167928865407559
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.35,
9
+ "acc_stderr": 0.015090650341444236
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.32,
13
+ "acc_stderr": 0.013471620929769152
14
+ },
15
+ "cb": {
16
+ "acc": 0.4107142857142857,
17
+ "acc_stderr": 0.0663363415035954,
18
+ "f1": 0.28651292802236195
19
+ },
20
+ "copa": {
21
+ "acc": 0.68,
22
+ "acc_stderr": 0.04688261722621504
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2930691097390958,
26
+ "acc_stderr": 0.004542396269999213,
27
+ "acc_norm": 0.3207528380800637,
28
+ "acc_norm_stderr": 0.004658120152230808
29
+ },
30
+ "rte": {
31
+ "acc": 0.5631768953068592,
32
+ "acc_stderr": 0.02985524739031495
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.4964483030781373,
36
+ "acc_stderr": 0.01405213114691586
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5793693212185996,
40
+ "acc_stderr": 0.01141582799434265
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5125382262996941,
44
+ "acc_stderr": 0.008742304974218311
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4313973063973064,
48
+ "acc_stderr": 0.010162752847747498,
49
+ "acc_norm": 0.38341750841750843,
50
+ "acc_norm_stderr": 0.009976995068264717
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19027303754266212,
54
+ "acc_stderr": 0.011470424179225709,
55
+ "acc_norm": 0.22610921501706485,
56
+ "acc_norm_stderr": 0.012224202097063274
57
+ },
58
+ "sciq": {
59
+ "acc": 0.704,
60
+ "acc_stderr": 0.014442734941575022,
61
+ "acc_norm": 0.658,
62
+ "acc_norm_stderr": 0.015008706182121731
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6240478781284005,
66
+ "acc_stderr": 0.011301098166895732,
67
+ "acc_norm": 0.6158868335146899,
68
+ "acc_norm_stderr": 0.011348160741479136
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_2.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.308,
5
+ "acc_stderr": 0.014606483127342763
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.353,
9
+ "acc_stderr": 0.015120172605483696
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3375,
13
+ "acc_stderr": 0.013655897185463653
14
+ },
15
+ "cb": {
16
+ "acc": 0.4642857142857143,
17
+ "acc_stderr": 0.06724777654937658,
18
+ "f1": 0.316548463356974
19
+ },
20
+ "copa": {
21
+ "acc": 0.61,
22
+ "acc_stderr": 0.04902071300001975
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2929695279824736,
26
+ "acc_stderr": 0.004541944342035901,
27
+ "acc_norm": 0.31866162119099783,
28
+ "acc_norm_stderr": 0.00465005215009441
29
+ },
30
+ "rte": {
31
+ "acc": 0.47653429602888087,
32
+ "acc_stderr": 0.030063300411902652
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5011838989739542,
36
+ "acc_stderr": 0.014052446290529012
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5836451095670764,
40
+ "acc_stderr": 0.011399490926937005
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4776758409785933,
44
+ "acc_stderr": 0.00873633411558504
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42634680134680136,
48
+ "acc_stderr": 0.010147858603835139,
49
+ "acc_norm": 0.3926767676767677,
50
+ "acc_norm_stderr": 0.010020646555538686
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.18344709897610922,
54
+ "acc_stderr": 0.011310170179554543,
55
+ "acc_norm": 0.22696245733788395,
56
+ "acc_norm_stderr": 0.012240491536132879
57
+ },
58
+ "sciq": {
59
+ "acc": 0.73,
60
+ "acc_stderr": 0.014046255632633915,
61
+ "acc_norm": 0.677,
62
+ "acc_norm_stderr": 0.014794927843348633
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6349292709466812,
66
+ "acc_stderr": 0.011233021830554829,
67
+ "acc_norm": 0.6251360174102285,
68
+ "acc_norm_stderr": 0.011294565805619019
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_2_lm-eval_global_step52452_2023-02-09-17-36-57_2shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.308,
5
+ "acc_stderr": 0.014606483127342763
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.353,
9
+ "acc_stderr": 0.015120172605483696
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3375,
13
+ "acc_stderr": 0.013655897185463653
14
+ },
15
+ "cb": {
16
+ "acc": 0.4642857142857143,
17
+ "acc_stderr": 0.06724777654937658,
18
+ "f1": 0.316548463356974
19
+ },
20
+ "copa": {
21
+ "acc": 0.61,
22
+ "acc_stderr": 0.04902071300001975
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2929695279824736,
26
+ "acc_stderr": 0.004541944342035901,
27
+ "acc_norm": 0.31866162119099783,
28
+ "acc_norm_stderr": 0.00465005215009441
29
+ },
30
+ "rte": {
31
+ "acc": 0.47653429602888087,
32
+ "acc_stderr": 0.030063300411902652
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5011838989739542,
36
+ "acc_stderr": 0.014052446290529012
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5836451095670764,
40
+ "acc_stderr": 0.011399490926937005
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4776758409785933,
44
+ "acc_stderr": 0.00873633411558504
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42634680134680136,
48
+ "acc_stderr": 0.010147858603835139,
49
+ "acc_norm": 0.3926767676767677,
50
+ "acc_norm_stderr": 0.010020646555538686
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.18344709897610922,
54
+ "acc_stderr": 0.011310170179554543,
55
+ "acc_norm": 0.22696245733788395,
56
+ "acc_norm_stderr": 0.012240491536132879
57
+ },
58
+ "sciq": {
59
+ "acc": 0.73,
60
+ "acc_stderr": 0.014046255632633915,
61
+ "acc_norm": 0.677,
62
+ "acc_norm_stderr": 0.014794927843348633
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6349292709466812,
66
+ "acc_stderr": 0.011233021830554829,
67
+ "acc_norm": 0.6251360174102285,
68
+ "acc_norm_stderr": 0.011294565805619019
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_3.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.314,
5
+ "acc_stderr": 0.014683991951087966
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.341,
9
+ "acc_stderr": 0.014998131348402704
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3466666666666667,
13
+ "acc_stderr": 0.013744022550571949
14
+ },
15
+ "cb": {
16
+ "acc": 0.5178571428571429,
17
+ "acc_stderr": 0.06737697508644648,
18
+ "f1": 0.347985347985348
19
+ },
20
+ "copa": {
21
+ "acc": 0.59,
22
+ "acc_stderr": 0.04943110704237101
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2953594901414061,
26
+ "acc_stderr": 0.004552718360513099,
27
+ "acc_norm": 0.3241386178052181,
28
+ "acc_norm_stderr": 0.0046709553996411276
29
+ },
30
+ "rte": {
31
+ "acc": 0.5090252707581228,
32
+ "acc_stderr": 0.030091559826331334
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.494869771112865,
36
+ "acc_stderr": 0.014051745961790513
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5788348476750401,
40
+ "acc_stderr": 0.011417808278216117
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4694189602446483,
44
+ "acc_stderr": 0.008728682900189723
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4276094276094276,
48
+ "acc_stderr": 0.010151683397430679,
49
+ "acc_norm": 0.39141414141414144,
50
+ "acc_norm_stderr": 0.010014917532627812
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19027303754266212,
54
+ "acc_stderr": 0.011470424179225698,
55
+ "acc_norm": 0.2235494880546075,
56
+ "acc_norm_stderr": 0.012174896631202607
57
+ },
58
+ "sciq": {
59
+ "acc": 0.716,
60
+ "acc_stderr": 0.014267009061031314,
61
+ "acc_norm": 0.679,
62
+ "acc_norm_stderr": 0.014770821817934645
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6289445048966268,
66
+ "acc_stderr": 0.011271222398600525,
67
+ "acc_norm": 0.6202393906420022,
68
+ "acc_norm_stderr": 0.011323483504715843
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_3_lm-eval_global_step52452_2023-02-09-17-36-57_3shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.314,
5
+ "acc_stderr": 0.014683991951087966
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.341,
9
+ "acc_stderr": 0.014998131348402704
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3466666666666667,
13
+ "acc_stderr": 0.013744022550571949
14
+ },
15
+ "cb": {
16
+ "acc": 0.5178571428571429,
17
+ "acc_stderr": 0.06737697508644648,
18
+ "f1": 0.347985347985348
19
+ },
20
+ "copa": {
21
+ "acc": 0.59,
22
+ "acc_stderr": 0.04943110704237101
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2953594901414061,
26
+ "acc_stderr": 0.004552718360513099,
27
+ "acc_norm": 0.3241386178052181,
28
+ "acc_norm_stderr": 0.0046709553996411276
29
+ },
30
+ "rte": {
31
+ "acc": 0.5090252707581228,
32
+ "acc_stderr": 0.030091559826331334
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.494869771112865,
36
+ "acc_stderr": 0.014051745961790513
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5788348476750401,
40
+ "acc_stderr": 0.011417808278216117
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4694189602446483,
44
+ "acc_stderr": 0.008728682900189723
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4276094276094276,
48
+ "acc_stderr": 0.010151683397430679,
49
+ "acc_norm": 0.39141414141414144,
50
+ "acc_norm_stderr": 0.010014917532627812
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19027303754266212,
54
+ "acc_stderr": 0.011470424179225698,
55
+ "acc_norm": 0.2235494880546075,
56
+ "acc_norm_stderr": 0.012174896631202607
57
+ },
58
+ "sciq": {
59
+ "acc": 0.716,
60
+ "acc_stderr": 0.014267009061031314,
61
+ "acc_norm": 0.679,
62
+ "acc_norm_stderr": 0.014770821817934645
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6289445048966268,
66
+ "acc_stderr": 0.011271222398600525,
67
+ "acc_norm": 0.6202393906420022,
68
+ "acc_norm_stderr": 0.011323483504715843
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_4.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.336,
5
+ "acc_stderr": 0.014944140233795025
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.353,
9
+ "acc_stderr": 0.01512017260548369
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3375,
13
+ "acc_stderr": 0.01365589718546366
14
+ },
15
+ "cb": {
16
+ "acc": 0.5178571428571429,
17
+ "acc_stderr": 0.06737697508644645,
18
+ "f1": 0.3362023995826813
19
+ },
20
+ "copa": {
21
+ "acc": 0.6,
22
+ "acc_stderr": 0.04923659639173309
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2938657637920733,
26
+ "acc_stderr": 0.004546002255456781,
27
+ "acc_norm": 0.32204740091615214,
28
+ "acc_norm_stderr": 0.00466306082837678
29
+ },
30
+ "rte": {
31
+ "acc": 0.49097472924187724,
32
+ "acc_stderr": 0.030091559826331334
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.48303078137332284,
36
+ "acc_stderr": 0.014044390401612969
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5793693212185996,
40
+ "acc_stderr": 0.011415827994342653
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4437308868501529,
44
+ "acc_stderr": 0.008689501105367405
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.43013468013468015,
48
+ "acc_stderr": 0.010159130445178514,
49
+ "acc_norm": 0.39225589225589225,
50
+ "acc_norm_stderr": 0.010018744689650043
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19795221843003413,
54
+ "acc_stderr": 0.011643990971573395,
55
+ "acc_norm": 0.23122866894197952,
56
+ "acc_norm_stderr": 0.012320858834772266
57
+ },
58
+ "sciq": {
59
+ "acc": 0.721,
60
+ "acc_stderr": 0.01419015011761203,
61
+ "acc_norm": 0.686,
62
+ "acc_norm_stderr": 0.014683991951087967
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6322089227421109,
66
+ "acc_stderr": 0.011250616646678797,
67
+ "acc_norm": 0.6240478781284005,
68
+ "acc_norm_stderr": 0.011301098166895724
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_4_lm-eval_global_step52452_2023-02-09-17-38-12_4shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.336,
5
+ "acc_stderr": 0.014944140233795025
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.353,
9
+ "acc_stderr": 0.01512017260548369
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3375,
13
+ "acc_stderr": 0.01365589718546366
14
+ },
15
+ "cb": {
16
+ "acc": 0.5178571428571429,
17
+ "acc_stderr": 0.06737697508644645,
18
+ "f1": 0.3362023995826813
19
+ },
20
+ "copa": {
21
+ "acc": 0.6,
22
+ "acc_stderr": 0.04923659639173309
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2938657637920733,
26
+ "acc_stderr": 0.004546002255456781,
27
+ "acc_norm": 0.32204740091615214,
28
+ "acc_norm_stderr": 0.00466306082837678
29
+ },
30
+ "rte": {
31
+ "acc": 0.49097472924187724,
32
+ "acc_stderr": 0.030091559826331334
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.48303078137332284,
36
+ "acc_stderr": 0.014044390401612969
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5793693212185996,
40
+ "acc_stderr": 0.011415827994342653
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4437308868501529,
44
+ "acc_stderr": 0.008689501105367405
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.43013468013468015,
48
+ "acc_stderr": 0.010159130445178514,
49
+ "acc_norm": 0.39225589225589225,
50
+ "acc_norm_stderr": 0.010018744689650043
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19795221843003413,
54
+ "acc_stderr": 0.011643990971573395,
55
+ "acc_norm": 0.23122866894197952,
56
+ "acc_norm_stderr": 0.012320858834772266
57
+ },
58
+ "sciq": {
59
+ "acc": 0.721,
60
+ "acc_stderr": 0.01419015011761203,
61
+ "acc_norm": 0.686,
62
+ "acc_norm_stderr": 0.014683991951087967
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6322089227421109,
66
+ "acc_stderr": 0.011250616646678797,
67
+ "acc_norm": 0.6240478781284005,
68
+ "acc_norm_stderr": 0.011301098166895724
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_5.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.368,
5
+ "acc_stderr": 0.015258073561521802
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.335,
9
+ "acc_stderr": 0.014933117490932577
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3416666666666667,
13
+ "acc_stderr": 0.013696658778002515
14
+ },
15
+ "cb": {
16
+ "acc": 0.5535714285714286,
17
+ "acc_stderr": 0.06703189227942395,
18
+ "f1": 0.3459575611066344
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145633
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2944632543318064,
26
+ "acc_stderr": 0.00454869574962096,
27
+ "acc_norm": 0.32423819956184025,
28
+ "acc_norm_stderr": 0.0046713286732178
29
+ },
30
+ "rte": {
31
+ "acc": 0.4981949458483754,
32
+ "acc_stderr": 0.030096267148976633
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.4972375690607735,
36
+ "acc_stderr": 0.014052271211616441
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5836451095670764,
40
+ "acc_stderr": 0.011399490926937006
41
+ },
42
+ "boolq": {
43
+ "acc": 0.44128440366972477,
44
+ "acc_stderr": 0.008684548127832634
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4297138047138047,
48
+ "acc_stderr": 0.010157908005763678,
49
+ "acc_norm": 0.39941077441077444,
50
+ "acc_norm_stderr": 0.010050018228742115
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19283276450511946,
54
+ "acc_stderr": 0.011529055465663338,
55
+ "acc_norm": 0.22696245733788395,
56
+ "acc_norm_stderr": 0.012240491536132873
57
+ },
58
+ "sciq": {
59
+ "acc": 0.717,
60
+ "acc_stderr": 0.014251810906481735,
61
+ "acc_norm": 0.68,
62
+ "acc_norm_stderr": 0.014758652303574883
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6278563656147987,
66
+ "acc_stderr": 0.01127796831359274,
67
+ "acc_norm": 0.6207834602829162,
68
+ "acc_norm_stderr": 0.011320331012905077
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_5_lm-eval_global_step52452_2023-02-09-17-38-12_5shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.368,
5
+ "acc_stderr": 0.015258073561521802
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.335,
9
+ "acc_stderr": 0.014933117490932577
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3416666666666667,
13
+ "acc_stderr": 0.013696658778002515
14
+ },
15
+ "cb": {
16
+ "acc": 0.5535714285714286,
17
+ "acc_stderr": 0.06703189227942395,
18
+ "f1": 0.3459575611066344
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145633
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2944632543318064,
26
+ "acc_stderr": 0.00454869574962096,
27
+ "acc_norm": 0.32423819956184025,
28
+ "acc_norm_stderr": 0.0046713286732178
29
+ },
30
+ "rte": {
31
+ "acc": 0.4981949458483754,
32
+ "acc_stderr": 0.030096267148976633
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.4972375690607735,
36
+ "acc_stderr": 0.014052271211616441
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5836451095670764,
40
+ "acc_stderr": 0.011399490926937006
41
+ },
42
+ "boolq": {
43
+ "acc": 0.44128440366972477,
44
+ "acc_stderr": 0.008684548127832634
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4297138047138047,
48
+ "acc_stderr": 0.010157908005763678,
49
+ "acc_norm": 0.39941077441077444,
50
+ "acc_norm_stderr": 0.010050018228742115
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19283276450511946,
54
+ "acc_stderr": 0.011529055465663338,
55
+ "acc_norm": 0.22696245733788395,
56
+ "acc_norm_stderr": 0.012240491536132873
57
+ },
58
+ "sciq": {
59
+ "acc": 0.717,
60
+ "acc_stderr": 0.014251810906481735,
61
+ "acc_norm": 0.68,
62
+ "acc_norm_stderr": 0.014758652303574883
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6278563656147987,
66
+ "acc_stderr": 0.01127796831359274,
67
+ "acc_norm": 0.6207834602829162,
68
+ "acc_norm_stderr": 0.011320331012905077
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_0.csv ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ task,metric,value,err,version
2
+ anli_r1,acc,0.333,0.014910846164229868,0
3
+ anli_r2,acc,0.339,0.014976758771620344,0
4
+ anli_r3,acc,0.33916666666666667,0.013672343491681819,0
5
+ arc_challenge,acc,0.18600682593856654,0.011370940183266749,0
6
+ arc_challenge,acc_norm,0.22610921501706485,0.01222420209706328,0
7
+ arc_easy,acc,0.42003367003367004,0.010127718838529398,0
8
+ arc_easy,acc_norm,0.3728956228956229,0.009922743197129255,0
9
+ boolq,acc,0.6051987767584098,0.008549304887647411,1
10
+ cb,acc,0.4107142857142857,0.0663363415035954,1
11
+ cb,f1,0.1940928270042194,,1
12
+ copa,acc,0.62,0.04878317312145632,0
13
+ hellaswag,acc,0.2949611631149173,0.004550933142528753,0
14
+ hellaswag,acc_norm,0.32463652658832903,0.004672819355838551,0
15
+ piqa,acc,0.6251360174102285,0.011294565805619017,0
16
+ piqa,acc_norm,0.6224156692056583,0.011310782787145772,0
17
+ rte,acc,0.5342960288808665,0.030025579819366422,0
18
+ sciq,acc,0.735,0.013963164754809949,0
19
+ sciq,acc_norm,0.656,0.015029633724408945,0
20
+ storycloze_2016,acc,0.5873864243719936,0.011384472322969045,0
21
+ winogrande,acc,0.516179952644041,0.014045126130978601,0
evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_0.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.333,
5
+ "acc_stderr": 0.014910846164229868
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.339,
9
+ "acc_stderr": 0.014976758771620344
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.33916666666666667,
13
+ "acc_stderr": 0.013672343491681819
14
+ },
15
+ "cb": {
16
+ "acc": 0.4107142857142857,
17
+ "acc_stderr": 0.0663363415035954,
18
+ "f1": 0.1940928270042194
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145632
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2949611631149173,
26
+ "acc_stderr": 0.004550933142528753,
27
+ "acc_norm": 0.32463652658832903,
28
+ "acc_norm_stderr": 0.004672819355838551
29
+ },
30
+ "rte": {
31
+ "acc": 0.5342960288808665,
32
+ "acc_stderr": 0.030025579819366422
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.516179952644041,
36
+ "acc_stderr": 0.014045126130978601
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5873864243719936,
40
+ "acc_stderr": 0.011384472322969045
41
+ },
42
+ "boolq": {
43
+ "acc": 0.6051987767584098,
44
+ "acc_stderr": 0.008549304887647411
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42003367003367004,
48
+ "acc_stderr": 0.010127718838529398,
49
+ "acc_norm": 0.3728956228956229,
50
+ "acc_norm_stderr": 0.009922743197129255
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.18600682593856654,
54
+ "acc_stderr": 0.011370940183266749,
55
+ "acc_norm": 0.22610921501706485,
56
+ "acc_norm_stderr": 0.01222420209706328
57
+ },
58
+ "sciq": {
59
+ "acc": 0.735,
60
+ "acc_stderr": 0.013963164754809949,
61
+ "acc_norm": 0.656,
62
+ "acc_norm_stderr": 0.015029633724408945
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6251360174102285,
66
+ "acc_stderr": 0.011294565805619017,
67
+ "acc_norm": 0.6224156692056583,
68
+ "acc_norm_stderr": 0.011310782787145772
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_0_lm-eval_global_step42000_2023-02-08-13-42-29_0shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.333,
5
+ "acc_stderr": 0.014910846164229868
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.339,
9
+ "acc_stderr": 0.014976758771620344
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.33916666666666667,
13
+ "acc_stderr": 0.013672343491681819
14
+ },
15
+ "cb": {
16
+ "acc": 0.4107142857142857,
17
+ "acc_stderr": 0.0663363415035954,
18
+ "f1": 0.1940928270042194
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145632
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2949611631149173,
26
+ "acc_stderr": 0.004550933142528753,
27
+ "acc_norm": 0.32463652658832903,
28
+ "acc_norm_stderr": 0.004672819355838551
29
+ },
30
+ "rte": {
31
+ "acc": 0.5342960288808665,
32
+ "acc_stderr": 0.030025579819366422
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.516179952644041,
36
+ "acc_stderr": 0.014045126130978601
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5873864243719936,
40
+ "acc_stderr": 0.011384472322969045
41
+ },
42
+ "boolq": {
43
+ "acc": 0.6051987767584098,
44
+ "acc_stderr": 0.008549304887647411
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42003367003367004,
48
+ "acc_stderr": 0.010127718838529398,
49
+ "acc_norm": 0.3728956228956229,
50
+ "acc_norm_stderr": 0.009922743197129255
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.18600682593856654,
54
+ "acc_stderr": 0.011370940183266749,
55
+ "acc_norm": 0.22610921501706485,
56
+ "acc_norm_stderr": 0.01222420209706328
57
+ },
58
+ "sciq": {
59
+ "acc": 0.735,
60
+ "acc_stderr": 0.013963164754809949,
61
+ "acc_norm": 0.656,
62
+ "acc_norm_stderr": 0.015029633724408945
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6251360174102285,
66
+ "acc_stderr": 0.011294565805619017,
67
+ "acc_norm": 0.6224156692056583,
68
+ "acc_norm_stderr": 0.011310782787145772
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_1.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.348,
5
+ "acc_stderr": 0.01507060460376841
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.335,
9
+ "acc_stderr": 0.014933117490932575
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3416666666666667,
13
+ "acc_stderr": 0.013696658778002505
14
+ },
15
+ "cb": {
16
+ "acc": 0.44642857142857145,
17
+ "acc_stderr": 0.06703189227942398,
18
+ "f1": 0.2712571726656234
19
+ },
20
+ "copa": {
21
+ "acc": 0.65,
22
+ "acc_stderr": 0.04793724854411019
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29396534554869547,
26
+ "acc_stderr": 0.004546451825028366,
27
+ "acc_norm": 0.3170683130850428,
28
+ "acc_norm_stderr": 0.004643832742876639
29
+ },
30
+ "rte": {
31
+ "acc": 0.5487364620938628,
32
+ "acc_stderr": 0.029953149241808946
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5067087608524072,
36
+ "acc_stderr": 0.014051220692330349
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5783003741314805,
40
+ "acc_stderr": 0.011419774841868156
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5587155963302752,
44
+ "acc_stderr": 0.008684548127832634
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.41203703703703703,
48
+ "acc_stderr": 0.010099765857562773,
49
+ "acc_norm": 0.3720538720538721,
50
+ "acc_norm_stderr": 0.009918187193096468
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.181740614334471,
54
+ "acc_stderr": 0.011269198948880236,
55
+ "acc_norm": 0.2167235494880546,
56
+ "acc_norm_stderr": 0.012040156713481192
57
+ },
58
+ "sciq": {
59
+ "acc": 0.685,
60
+ "acc_stderr": 0.014696631960792492,
61
+ "acc_norm": 0.632,
62
+ "acc_norm_stderr": 0.0152580735615218
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6294885745375408,
66
+ "acc_stderr": 0.011267826475447665,
67
+ "acc_norm": 0.6262241566920566,
68
+ "acc_norm_stderr": 0.011287972563201017
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_1_lm-eval_global_step42000_2023-02-08-13-42-29_1shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.348,
5
+ "acc_stderr": 0.01507060460376841
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.335,
9
+ "acc_stderr": 0.014933117490932575
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3416666666666667,
13
+ "acc_stderr": 0.013696658778002505
14
+ },
15
+ "cb": {
16
+ "acc": 0.44642857142857145,
17
+ "acc_stderr": 0.06703189227942398,
18
+ "f1": 0.2712571726656234
19
+ },
20
+ "copa": {
21
+ "acc": 0.65,
22
+ "acc_stderr": 0.04793724854411019
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29396534554869547,
26
+ "acc_stderr": 0.004546451825028366,
27
+ "acc_norm": 0.3170683130850428,
28
+ "acc_norm_stderr": 0.004643832742876639
29
+ },
30
+ "rte": {
31
+ "acc": 0.5487364620938628,
32
+ "acc_stderr": 0.029953149241808946
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5067087608524072,
36
+ "acc_stderr": 0.014051220692330349
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5783003741314805,
40
+ "acc_stderr": 0.011419774841868156
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5587155963302752,
44
+ "acc_stderr": 0.008684548127832634
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.41203703703703703,
48
+ "acc_stderr": 0.010099765857562773,
49
+ "acc_norm": 0.3720538720538721,
50
+ "acc_norm_stderr": 0.009918187193096468
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.181740614334471,
54
+ "acc_stderr": 0.011269198948880236,
55
+ "acc_norm": 0.2167235494880546,
56
+ "acc_norm_stderr": 0.012040156713481192
57
+ },
58
+ "sciq": {
59
+ "acc": 0.685,
60
+ "acc_stderr": 0.014696631960792492,
61
+ "acc_norm": 0.632,
62
+ "acc_norm_stderr": 0.0152580735615218
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6294885745375408,
66
+ "acc_stderr": 0.011267826475447665,
67
+ "acc_norm": 0.6262241566920566,
68
+ "acc_norm_stderr": 0.011287972563201017
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_2.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.313,
5
+ "acc_stderr": 0.014671272822977892
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.336,
9
+ "acc_stderr": 0.014944140233795023
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3425,
13
+ "acc_stderr": 0.013704669762934732
14
+ },
15
+ "cb": {
16
+ "acc": 0.42857142857142855,
17
+ "acc_stderr": 0.06672848092813058,
18
+ "f1": 0.2791044776119403
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145633
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2887870942043418,
26
+ "acc_stderr": 0.004522725412556968,
27
+ "acc_norm": 0.31517625970922125,
28
+ "acc_norm_stderr": 0.004636365534819762
29
+ },
30
+ "rte": {
31
+ "acc": 0.48014440433212996,
32
+ "acc_stderr": 0.0300727231673172
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5011838989739542,
36
+ "acc_stderr": 0.014052446290529012
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5809727418492785,
40
+ "acc_stderr": 0.011409804749706194
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5483180428134556,
44
+ "acc_stderr": 0.008704126206159355
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.398989898989899,
48
+ "acc_stderr": 0.010048240683798759,
49
+ "acc_norm": 0.36784511784511786,
50
+ "acc_norm_stderr": 0.009894923464455196
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.18515358361774745,
54
+ "acc_stderr": 0.011350774438389695,
55
+ "acc_norm": 0.22781569965870307,
56
+ "acc_norm_stderr": 0.012256708602326914
57
+ },
58
+ "sciq": {
59
+ "acc": 0.717,
60
+ "acc_stderr": 0.014251810906481735,
61
+ "acc_norm": 0.634,
62
+ "acc_norm_stderr": 0.015240612726405756
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6327529923830251,
66
+ "acc_stderr": 0.011247128539690563,
67
+ "acc_norm": 0.6175190424374319,
68
+ "acc_norm_stderr": 0.011339019654272345
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_2_lm-eval_global_step42000_2023-02-08-13-42-29_2shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.313,
5
+ "acc_stderr": 0.014671272822977892
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.336,
9
+ "acc_stderr": 0.014944140233795023
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3425,
13
+ "acc_stderr": 0.013704669762934732
14
+ },
15
+ "cb": {
16
+ "acc": 0.42857142857142855,
17
+ "acc_stderr": 0.06672848092813058,
18
+ "f1": 0.2791044776119403
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145633
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2887870942043418,
26
+ "acc_stderr": 0.004522725412556968,
27
+ "acc_norm": 0.31517625970922125,
28
+ "acc_norm_stderr": 0.004636365534819762
29
+ },
30
+ "rte": {
31
+ "acc": 0.48014440433212996,
32
+ "acc_stderr": 0.0300727231673172
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5011838989739542,
36
+ "acc_stderr": 0.014052446290529012
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5809727418492785,
40
+ "acc_stderr": 0.011409804749706194
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5483180428134556,
44
+ "acc_stderr": 0.008704126206159355
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.398989898989899,
48
+ "acc_stderr": 0.010048240683798759,
49
+ "acc_norm": 0.36784511784511786,
50
+ "acc_norm_stderr": 0.009894923464455196
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.18515358361774745,
54
+ "acc_stderr": 0.011350774438389695,
55
+ "acc_norm": 0.22781569965870307,
56
+ "acc_norm_stderr": 0.012256708602326914
57
+ },
58
+ "sciq": {
59
+ "acc": 0.717,
60
+ "acc_stderr": 0.014251810906481735,
61
+ "acc_norm": 0.634,
62
+ "acc_norm_stderr": 0.015240612726405756
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6327529923830251,
66
+ "acc_stderr": 0.011247128539690563,
67
+ "acc_norm": 0.6175190424374319,
68
+ "acc_norm_stderr": 0.011339019654272345
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_3.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.321,
5
+ "acc_stderr": 0.01477082181793465
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.343,
9
+ "acc_stderr": 0.015019206922356953
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.33916666666666667,
13
+ "acc_stderr": 0.013672343491681822
14
+ },
15
+ "cb": {
16
+ "acc": 0.39285714285714285,
17
+ "acc_stderr": 0.0658538889806635,
18
+ "f1": 0.2593406593406593
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145633
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29087831109340767,
26
+ "acc_stderr": 0.004532393111248679,
27
+ "acc_norm": 0.3136825333598885,
28
+ "acc_norm_stderr": 0.004630407476835188
29
+ },
30
+ "rte": {
31
+ "acc": 0.5090252707581228,
32
+ "acc_stderr": 0.030091559826331334
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5146014206787688,
36
+ "acc_stderr": 0.014046492383275835
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5783003741314805,
40
+ "acc_stderr": 0.011419774841868156
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5495412844036697,
44
+ "acc_stderr": 0.008702022442950878
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4090909090909091,
48
+ "acc_stderr": 0.010088775152615779,
49
+ "acc_norm": 0.3686868686868687,
50
+ "acc_norm_stderr": 0.009899640855681038
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.18771331058020477,
54
+ "acc_stderr": 0.011411001314155136,
55
+ "acc_norm": 0.22098976109215018,
56
+ "acc_norm_stderr": 0.012124929206818258
57
+ },
58
+ "sciq": {
59
+ "acc": 0.694,
60
+ "acc_stderr": 0.014580006055436972,
61
+ "acc_norm": 0.652,
62
+ "acc_norm_stderr": 0.015070604603768408
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6354733405875952,
66
+ "acc_stderr": 0.011229456510295966,
67
+ "acc_norm": 0.6262241566920566,
68
+ "acc_norm_stderr": 0.011287972563201014
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_3_lm-eval_global_step42000_2023-02-08-13-42-29_3shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.321,
5
+ "acc_stderr": 0.01477082181793465
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.343,
9
+ "acc_stderr": 0.015019206922356953
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.33916666666666667,
13
+ "acc_stderr": 0.013672343491681822
14
+ },
15
+ "cb": {
16
+ "acc": 0.39285714285714285,
17
+ "acc_stderr": 0.0658538889806635,
18
+ "f1": 0.2593406593406593
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145633
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29087831109340767,
26
+ "acc_stderr": 0.004532393111248679,
27
+ "acc_norm": 0.3136825333598885,
28
+ "acc_norm_stderr": 0.004630407476835188
29
+ },
30
+ "rte": {
31
+ "acc": 0.5090252707581228,
32
+ "acc_stderr": 0.030091559826331334
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5146014206787688,
36
+ "acc_stderr": 0.014046492383275835
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5783003741314805,
40
+ "acc_stderr": 0.011419774841868156
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5495412844036697,
44
+ "acc_stderr": 0.008702022442950878
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4090909090909091,
48
+ "acc_stderr": 0.010088775152615779,
49
+ "acc_norm": 0.3686868686868687,
50
+ "acc_norm_stderr": 0.009899640855681038
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.18771331058020477,
54
+ "acc_stderr": 0.011411001314155136,
55
+ "acc_norm": 0.22098976109215018,
56
+ "acc_norm_stderr": 0.012124929206818258
57
+ },
58
+ "sciq": {
59
+ "acc": 0.694,
60
+ "acc_stderr": 0.014580006055436972,
61
+ "acc_norm": 0.652,
62
+ "acc_norm_stderr": 0.015070604603768408
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6354733405875952,
66
+ "acc_stderr": 0.011229456510295966,
67
+ "acc_norm": 0.6262241566920566,
68
+ "acc_norm_stderr": 0.011287972563201014
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_4.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.34,
5
+ "acc_stderr": 0.014987482264363937
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.341,
9
+ "acc_stderr": 0.014998131348402697
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.355,
13
+ "acc_stderr": 0.013819249004047298
14
+ },
15
+ "cb": {
16
+ "acc": 0.44642857142857145,
17
+ "acc_stderr": 0.06703189227942397,
18
+ "f1": 0.29572649572649573
19
+ },
20
+ "copa": {
21
+ "acc": 0.57,
22
+ "acc_stderr": 0.04975698519562428
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.287293367855009,
26
+ "acc_stderr": 0.004515748192605717,
27
+ "acc_norm": 0.3125871340370444,
28
+ "acc_norm_stderr": 0.004626002828389158
29
+ },
30
+ "rte": {
31
+ "acc": 0.4657039711191336,
32
+ "acc_stderr": 0.030025579819366426
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5169692186266772,
36
+ "acc_stderr": 0.014044390401612976
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5750935328701229,
40
+ "acc_stderr": 0.011431286492205843
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5513761467889908,
44
+ "acc_stderr": 0.008698767182005272
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4057239057239057,
48
+ "acc_stderr": 0.010075755540128876,
49
+ "acc_norm": 0.37626262626262624,
50
+ "acc_norm_stderr": 0.009940646221513786
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.1962457337883959,
54
+ "acc_stderr": 0.011606019881416286,
55
+ "acc_norm": 0.22781569965870307,
56
+ "acc_norm_stderr": 0.012256708602326905
57
+ },
58
+ "sciq": {
59
+ "acc": 0.691,
60
+ "acc_stderr": 0.014619600977206486,
61
+ "acc_norm": 0.658,
62
+ "acc_norm_stderr": 0.01500870618212173
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6305767138193689,
66
+ "acc_stderr": 0.011260988628572341,
67
+ "acc_norm": 0.6175190424374319,
68
+ "acc_norm_stderr": 0.011339019654272347
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_4_lm-eval_global_step42000_2023-02-08-13-42-29_4shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.34,
5
+ "acc_stderr": 0.014987482264363937
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.341,
9
+ "acc_stderr": 0.014998131348402697
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.355,
13
+ "acc_stderr": 0.013819249004047298
14
+ },
15
+ "cb": {
16
+ "acc": 0.44642857142857145,
17
+ "acc_stderr": 0.06703189227942397,
18
+ "f1": 0.29572649572649573
19
+ },
20
+ "copa": {
21
+ "acc": 0.57,
22
+ "acc_stderr": 0.04975698519562428
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.287293367855009,
26
+ "acc_stderr": 0.004515748192605717,
27
+ "acc_norm": 0.3125871340370444,
28
+ "acc_norm_stderr": 0.004626002828389158
29
+ },
30
+ "rte": {
31
+ "acc": 0.4657039711191336,
32
+ "acc_stderr": 0.030025579819366426
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5169692186266772,
36
+ "acc_stderr": 0.014044390401612976
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5750935328701229,
40
+ "acc_stderr": 0.011431286492205843
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5513761467889908,
44
+ "acc_stderr": 0.008698767182005272
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4057239057239057,
48
+ "acc_stderr": 0.010075755540128876,
49
+ "acc_norm": 0.37626262626262624,
50
+ "acc_norm_stderr": 0.009940646221513786
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.1962457337883959,
54
+ "acc_stderr": 0.011606019881416286,
55
+ "acc_norm": 0.22781569965870307,
56
+ "acc_norm_stderr": 0.012256708602326905
57
+ },
58
+ "sciq": {
59
+ "acc": 0.691,
60
+ "acc_stderr": 0.014619600977206486,
61
+ "acc_norm": 0.658,
62
+ "acc_norm_stderr": 0.01500870618212173
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6305767138193689,
66
+ "acc_stderr": 0.011260988628572341,
67
+ "acc_norm": 0.6175190424374319,
68
+ "acc_norm_stderr": 0.011339019654272347
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_5.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.353,
5
+ "acc_stderr": 0.01512017260548369
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.342,
9
+ "acc_stderr": 0.01500870618212173
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3575,
13
+ "acc_stderr": 0.013840921245257794
14
+ },
15
+ "cb": {
16
+ "acc": 0.5,
17
+ "acc_stderr": 0.06741998624632421,
18
+ "f1": 0.34521263958184845
19
+ },
20
+ "copa": {
21
+ "acc": 0.56,
22
+ "acc_stderr": 0.04988876515698589
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2891854212308305,
26
+ "acc_stderr": 0.004524575892952968,
27
+ "acc_norm": 0.3157737502489544,
28
+ "acc_norm_stderr": 0.004638733202373885
29
+ },
30
+ "rte": {
31
+ "acc": 0.5379061371841155,
32
+ "acc_stderr": 0.030009848912529117
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5019731649565904,
36
+ "acc_stderr": 0.014052376259225632
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5793693212185996,
40
+ "acc_stderr": 0.011415827994342657
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5370030581039755,
44
+ "acc_stderr": 0.008721074177479658
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4031986531986532,
48
+ "acc_stderr": 0.010065668576794787,
49
+ "acc_norm": 0.37457912457912457,
50
+ "acc_norm_stderr": 0.00993175882041061
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.189419795221843,
54
+ "acc_stderr": 0.01145070511591077,
55
+ "acc_norm": 0.22696245733788395,
56
+ "acc_norm_stderr": 0.012240491536132873
57
+ },
58
+ "sciq": {
59
+ "acc": 0.71,
60
+ "acc_stderr": 0.01435639599990569,
61
+ "acc_norm": 0.665,
62
+ "acc_norm_stderr": 0.014933117490932572
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6229597388465724,
66
+ "acc_stderr": 0.011307569752543902,
67
+ "acc_norm": 0.6126224156692056,
68
+ "acc_norm_stderr": 0.011366038083435908
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_s_denoiser_44b/checkpoints_2b855b55bc4ul2ndfixnew_5_lm-eval_global_step42000_2023-02-08-13-42-29_5shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.353,
5
+ "acc_stderr": 0.01512017260548369
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.342,
9
+ "acc_stderr": 0.01500870618212173
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3575,
13
+ "acc_stderr": 0.013840921245257794
14
+ },
15
+ "cb": {
16
+ "acc": 0.5,
17
+ "acc_stderr": 0.06741998624632421,
18
+ "f1": 0.34521263958184845
19
+ },
20
+ "copa": {
21
+ "acc": 0.56,
22
+ "acc_stderr": 0.04988876515698589
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2891854212308305,
26
+ "acc_stderr": 0.004524575892952968,
27
+ "acc_norm": 0.3157737502489544,
28
+ "acc_norm_stderr": 0.004638733202373885
29
+ },
30
+ "rte": {
31
+ "acc": 0.5379061371841155,
32
+ "acc_stderr": 0.030009848912529117
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5019731649565904,
36
+ "acc_stderr": 0.014052376259225632
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5793693212185996,
40
+ "acc_stderr": 0.011415827994342657
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5370030581039755,
44
+ "acc_stderr": 0.008721074177479658
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4031986531986532,
48
+ "acc_stderr": 0.010065668576794787,
49
+ "acc_norm": 0.37457912457912457,
50
+ "acc_norm_stderr": 0.00993175882041061
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.189419795221843,
54
+ "acc_stderr": 0.01145070511591077,
55
+ "acc_norm": 0.22696245733788395,
56
+ "acc_norm_stderr": 0.012240491536132873
57
+ },
58
+ "sciq": {
59
+ "acc": 0.71,
60
+ "acc_stderr": 0.01435639599990569,
61
+ "acc_norm": 0.665,
62
+ "acc_norm_stderr": 0.014933117490932572
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6229597388465724,
66
+ "acc_stderr": 0.011307569752543902,
67
+ "acc_norm": 0.6126224156692056,
68
+ "acc_norm_stderr": 0.011366038083435908
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0.csv ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ task,metric,value,err,version
2
+ anli_r1,acc,0.338,0.014965960710224473,0
3
+ anli_r2,acc,0.333,0.01491084616422986,0
4
+ anli_r3,acc,0.335,0.01363087184382148,0
5
+ arc_challenge,acc,0.181740614334471,0.011269198948880236,0
6
+ arc_challenge,acc_norm,0.22098976109215018,0.012124929206818258,0
7
+ arc_easy,acc,0.43434343434343436,0.010170943451269425,0
8
+ arc_easy,acc_norm,0.382996632996633,0.009974920384536472,0
9
+ boolq,acc,0.5935779816513761,0.008590531708882188,1
10
+ cb,acc,0.4107142857142857,0.0663363415035954,1
11
+ cb,f1,0.1940928270042194,,1
12
+ copa,acc,0.63,0.048523658709391,0
13
+ hellaswag,acc,0.29725154351722766,0.004561141293448468,0
14
+ hellaswag,acc_norm,0.3256323441545509,0.004676529200753,0
15
+ piqa,acc,0.6300326441784548,0.011264415223415281,0
16
+ piqa,acc_norm,0.6322089227421109,0.011250616646678792,0
17
+ rte,acc,0.5234657039711191,0.03006330041190266,0
18
+ sciq,acc,0.736,0.013946271849440472,0
19
+ sciq,acc_norm,0.668,0.014899597242811476,0
20
+ storycloze_2016,acc,0.5916622127204704,0.011366477562142522,0
21
+ winogrande,acc,0.5090765588003157,0.01405017009449771,0
evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.338,
5
+ "acc_stderr": 0.014965960710224473
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.333,
9
+ "acc_stderr": 0.01491084616422986
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.335,
13
+ "acc_stderr": 0.01363087184382148
14
+ },
15
+ "cb": {
16
+ "acc": 0.4107142857142857,
17
+ "acc_stderr": 0.0663363415035954,
18
+ "f1": 0.1940928270042194
19
+ },
20
+ "copa": {
21
+ "acc": 0.63,
22
+ "acc_stderr": 0.048523658709391
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29725154351722766,
26
+ "acc_stderr": 0.004561141293448468,
27
+ "acc_norm": 0.3256323441545509,
28
+ "acc_norm_stderr": 0.004676529200753
29
+ },
30
+ "rte": {
31
+ "acc": 0.5234657039711191,
32
+ "acc_stderr": 0.03006330041190266
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5090765588003157,
36
+ "acc_stderr": 0.01405017009449771
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5916622127204704,
40
+ "acc_stderr": 0.011366477562142522
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5935779816513761,
44
+ "acc_stderr": 0.008590531708882188
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.43434343434343436,
48
+ "acc_stderr": 0.010170943451269425,
49
+ "acc_norm": 0.382996632996633,
50
+ "acc_norm_stderr": 0.009974920384536472
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.181740614334471,
54
+ "acc_stderr": 0.011269198948880236,
55
+ "acc_norm": 0.22098976109215018,
56
+ "acc_norm_stderr": 0.012124929206818258
57
+ },
58
+ "sciq": {
59
+ "acc": 0.736,
60
+ "acc_stderr": 0.013946271849440472,
61
+ "acc_norm": 0.668,
62
+ "acc_norm_stderr": 0.014899597242811476
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6300326441784548,
66
+ "acc_stderr": 0.011264415223415281,
67
+ "acc_norm": 0.6322089227421109,
68
+ "acc_norm_stderr": 0.011250616646678792
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_0_lm-eval_global_step52452_2023-02-09-23-08-31_0shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.338,
5
+ "acc_stderr": 0.014965960710224473
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.333,
9
+ "acc_stderr": 0.01491084616422986
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.335,
13
+ "acc_stderr": 0.01363087184382148
14
+ },
15
+ "cb": {
16
+ "acc": 0.4107142857142857,
17
+ "acc_stderr": 0.0663363415035954,
18
+ "f1": 0.1940928270042194
19
+ },
20
+ "copa": {
21
+ "acc": 0.63,
22
+ "acc_stderr": 0.048523658709391
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29725154351722766,
26
+ "acc_stderr": 0.004561141293448468,
27
+ "acc_norm": 0.3256323441545509,
28
+ "acc_norm_stderr": 0.004676529200753
29
+ },
30
+ "rte": {
31
+ "acc": 0.5234657039711191,
32
+ "acc_stderr": 0.03006330041190266
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5090765588003157,
36
+ "acc_stderr": 0.01405017009449771
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5916622127204704,
40
+ "acc_stderr": 0.011366477562142522
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5935779816513761,
44
+ "acc_stderr": 0.008590531708882188
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.43434343434343436,
48
+ "acc_stderr": 0.010170943451269425,
49
+ "acc_norm": 0.382996632996633,
50
+ "acc_norm_stderr": 0.009974920384536472
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.181740614334471,
54
+ "acc_stderr": 0.011269198948880236,
55
+ "acc_norm": 0.22098976109215018,
56
+ "acc_norm_stderr": 0.012124929206818258
57
+ },
58
+ "sciq": {
59
+ "acc": 0.736,
60
+ "acc_stderr": 0.013946271849440472,
61
+ "acc_norm": 0.668,
62
+ "acc_norm_stderr": 0.014899597242811476
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6300326441784548,
66
+ "acc_stderr": 0.011264415223415281,
67
+ "acc_norm": 0.6322089227421109,
68
+ "acc_norm_stderr": 0.011250616646678792
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_1.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.359,
5
+ "acc_stderr": 0.015177264224798594
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.349,
9
+ "acc_stderr": 0.015080663991563102
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.32,
13
+ "acc_stderr": 0.01347162092976915
14
+ },
15
+ "cb": {
16
+ "acc": 0.39285714285714285,
17
+ "acc_stderr": 0.0658538889806635,
18
+ "f1": 0.27365967365967364
19
+ },
20
+ "copa": {
21
+ "acc": 0.64,
22
+ "acc_stderr": 0.048241815132442176
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2909778928500299,
26
+ "acc_stderr": 0.004532850566893522,
27
+ "acc_norm": 0.31955785700059747,
28
+ "acc_norm_stderr": 0.004653523038369371
29
+ },
30
+ "rte": {
31
+ "acc": 0.5523465703971119,
32
+ "acc_stderr": 0.02993107036293953
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5019731649565904,
36
+ "acc_stderr": 0.014052376259225632
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5783003741314805,
40
+ "acc_stderr": 0.011419774841868156
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5122324159021406,
44
+ "acc_stderr": 0.008742437504570405
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.43097643097643096,
48
+ "acc_stderr": 0.010161552863493744,
49
+ "acc_norm": 0.3792087542087542,
50
+ "acc_norm_stderr": 0.00995589166886556
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.1885665529010239,
54
+ "acc_stderr": 0.0114308976476758,
55
+ "acc_norm": 0.2158703071672355,
56
+ "acc_norm_stderr": 0.012022975360030668
57
+ },
58
+ "sciq": {
59
+ "acc": 0.703,
60
+ "acc_stderr": 0.0144568322948011,
61
+ "acc_norm": 0.659,
62
+ "acc_norm_stderr": 0.014998131348402706
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6224156692056583,
66
+ "acc_stderr": 0.011310782787145781,
67
+ "acc_norm": 0.6158868335146899,
68
+ "acc_norm_stderr": 0.011348160741479136
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_1_lm-eval_global_step52452_2023-02-09-23-08-31_1shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.359,
5
+ "acc_stderr": 0.015177264224798594
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.349,
9
+ "acc_stderr": 0.015080663991563102
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.32,
13
+ "acc_stderr": 0.01347162092976915
14
+ },
15
+ "cb": {
16
+ "acc": 0.39285714285714285,
17
+ "acc_stderr": 0.0658538889806635,
18
+ "f1": 0.27365967365967364
19
+ },
20
+ "copa": {
21
+ "acc": 0.64,
22
+ "acc_stderr": 0.048241815132442176
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2909778928500299,
26
+ "acc_stderr": 0.004532850566893522,
27
+ "acc_norm": 0.31955785700059747,
28
+ "acc_norm_stderr": 0.004653523038369371
29
+ },
30
+ "rte": {
31
+ "acc": 0.5523465703971119,
32
+ "acc_stderr": 0.02993107036293953
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5019731649565904,
36
+ "acc_stderr": 0.014052376259225632
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5783003741314805,
40
+ "acc_stderr": 0.011419774841868156
41
+ },
42
+ "boolq": {
43
+ "acc": 0.5122324159021406,
44
+ "acc_stderr": 0.008742437504570405
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.43097643097643096,
48
+ "acc_stderr": 0.010161552863493744,
49
+ "acc_norm": 0.3792087542087542,
50
+ "acc_norm_stderr": 0.00995589166886556
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.1885665529010239,
54
+ "acc_stderr": 0.0114308976476758,
55
+ "acc_norm": 0.2158703071672355,
56
+ "acc_norm_stderr": 0.012022975360030668
57
+ },
58
+ "sciq": {
59
+ "acc": 0.703,
60
+ "acc_stderr": 0.0144568322948011,
61
+ "acc_norm": 0.659,
62
+ "acc_norm_stderr": 0.014998131348402706
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6224156692056583,
66
+ "acc_stderr": 0.011310782787145781,
67
+ "acc_norm": 0.6158868335146899,
68
+ "acc_norm_stderr": 0.011348160741479136
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_2.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.312,
5
+ "acc_stderr": 0.014658474370509001
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.348,
9
+ "acc_stderr": 0.01507060460376841
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3358333333333333,
13
+ "acc_stderr": 0.013639261190932889
14
+ },
15
+ "cb": {
16
+ "acc": 0.44642857142857145,
17
+ "acc_stderr": 0.06703189227942398,
18
+ "f1": 0.30579096045197734
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145633
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29286994622585144,
26
+ "acc_stderr": 0.004541492151639241,
27
+ "acc_norm": 0.31736705835490936,
28
+ "acc_norm_stderr": 0.004645003662067885
29
+ },
30
+ "rte": {
31
+ "acc": 0.48375451263537905,
32
+ "acc_stderr": 0.030080573208738064
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5082872928176796,
36
+ "acc_stderr": 0.014050555322824189
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5793693212185996,
40
+ "acc_stderr": 0.011415827994342657
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4709480122324159,
44
+ "acc_stderr": 0.008730280528451546
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4212962962962963,
48
+ "acc_stderr": 0.010131882498193131,
49
+ "acc_norm": 0.3888888888888889,
50
+ "acc_norm_stderr": 0.010003248335313755
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.181740614334471,
54
+ "acc_stderr": 0.011269198948880236,
55
+ "acc_norm": 0.22866894197952217,
56
+ "acc_norm_stderr": 0.0122728535825408
57
+ },
58
+ "sciq": {
59
+ "acc": 0.727,
60
+ "acc_stderr": 0.014095022868717605,
61
+ "acc_norm": 0.676,
62
+ "acc_norm_stderr": 0.01480686473373886
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6349292709466812,
66
+ "acc_stderr": 0.011233021830554826,
67
+ "acc_norm": 0.6240478781284005,
68
+ "acc_norm_stderr": 0.011301098166895727
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_2_lm-eval_global_step52452_2023-02-09-23-08-31_2shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.312,
5
+ "acc_stderr": 0.014658474370509001
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.348,
9
+ "acc_stderr": 0.01507060460376841
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3358333333333333,
13
+ "acc_stderr": 0.013639261190932889
14
+ },
15
+ "cb": {
16
+ "acc": 0.44642857142857145,
17
+ "acc_stderr": 0.06703189227942398,
18
+ "f1": 0.30579096045197734
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145633
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29286994622585144,
26
+ "acc_stderr": 0.004541492151639241,
27
+ "acc_norm": 0.31736705835490936,
28
+ "acc_norm_stderr": 0.004645003662067885
29
+ },
30
+ "rte": {
31
+ "acc": 0.48375451263537905,
32
+ "acc_stderr": 0.030080573208738064
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.5082872928176796,
36
+ "acc_stderr": 0.014050555322824189
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5793693212185996,
40
+ "acc_stderr": 0.011415827994342657
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4709480122324159,
44
+ "acc_stderr": 0.008730280528451546
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.4212962962962963,
48
+ "acc_stderr": 0.010131882498193131,
49
+ "acc_norm": 0.3888888888888889,
50
+ "acc_norm_stderr": 0.010003248335313755
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.181740614334471,
54
+ "acc_stderr": 0.011269198948880236,
55
+ "acc_norm": 0.22866894197952217,
56
+ "acc_norm_stderr": 0.0122728535825408
57
+ },
58
+ "sciq": {
59
+ "acc": 0.727,
60
+ "acc_stderr": 0.014095022868717605,
61
+ "acc_norm": 0.676,
62
+ "acc_norm_stderr": 0.01480686473373886
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6349292709466812,
66
+ "acc_stderr": 0.011233021830554826,
67
+ "acc_norm": 0.6240478781284005,
68
+ "acc_norm_stderr": 0.011301098166895727
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_3.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.314,
5
+ "acc_stderr": 0.014683991951087967
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.343,
9
+ "acc_stderr": 0.015019206922356951
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3458333333333333,
13
+ "acc_stderr": 0.013736245342311012
14
+ },
15
+ "cb": {
16
+ "acc": 0.48214285714285715,
17
+ "acc_stderr": 0.0673769750864465,
18
+ "f1": 0.3218559218559219
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145633
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29286994622585144,
26
+ "acc_stderr": 0.004541492151639243,
27
+ "acc_norm": 0.3227444732125075,
28
+ "acc_norm_stderr": 0.004665704208339039
29
+ },
30
+ "rte": {
31
+ "acc": 0.5054151624548736,
32
+ "acc_stderr": 0.030094698123239966
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.4925019731649566,
36
+ "acc_stderr": 0.01405090552122858
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5729556386958845,
40
+ "acc_stderr": 0.01143868739857839
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4675840978593272,
44
+ "acc_stderr": 0.008726657178723137
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42297979797979796,
48
+ "acc_stderr": 0.010137328382209099,
49
+ "acc_norm": 0.3808922558922559,
50
+ "acc_norm_stderr": 0.009964428212260379
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19197952218430034,
54
+ "acc_stderr": 0.011509598906598098,
55
+ "acc_norm": 0.22525597269624573,
56
+ "acc_norm_stderr": 0.0122078399954073
57
+ },
58
+ "sciq": {
59
+ "acc": 0.72,
60
+ "acc_stderr": 0.014205696104091496,
61
+ "acc_norm": 0.679,
62
+ "acc_norm_stderr": 0.014770821817934645
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6273122959738846,
66
+ "acc_stderr": 0.011281318332897734,
67
+ "acc_norm": 0.6202393906420022,
68
+ "acc_norm_stderr": 0.01132348350471584
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_3_lm-eval_global_step52452_2023-02-09-23-08-31_3shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.314,
5
+ "acc_stderr": 0.014683991951087967
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.343,
9
+ "acc_stderr": 0.015019206922356951
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.3458333333333333,
13
+ "acc_stderr": 0.013736245342311012
14
+ },
15
+ "cb": {
16
+ "acc": 0.48214285714285715,
17
+ "acc_stderr": 0.0673769750864465,
18
+ "f1": 0.3218559218559219
19
+ },
20
+ "copa": {
21
+ "acc": 0.62,
22
+ "acc_stderr": 0.04878317312145633
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.29286994622585144,
26
+ "acc_stderr": 0.004541492151639243,
27
+ "acc_norm": 0.3227444732125075,
28
+ "acc_norm_stderr": 0.004665704208339039
29
+ },
30
+ "rte": {
31
+ "acc": 0.5054151624548736,
32
+ "acc_stderr": 0.030094698123239966
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.4925019731649566,
36
+ "acc_stderr": 0.01405090552122858
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5729556386958845,
40
+ "acc_stderr": 0.01143868739857839
41
+ },
42
+ "boolq": {
43
+ "acc": 0.4675840978593272,
44
+ "acc_stderr": 0.008726657178723137
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42297979797979796,
48
+ "acc_stderr": 0.010137328382209099,
49
+ "acc_norm": 0.3808922558922559,
50
+ "acc_norm_stderr": 0.009964428212260379
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19197952218430034,
54
+ "acc_stderr": 0.011509598906598098,
55
+ "acc_norm": 0.22525597269624573,
56
+ "acc_norm_stderr": 0.0122078399954073
57
+ },
58
+ "sciq": {
59
+ "acc": 0.72,
60
+ "acc_stderr": 0.014205696104091496,
61
+ "acc_norm": 0.679,
62
+ "acc_norm_stderr": 0.014770821817934645
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6273122959738846,
66
+ "acc_stderr": 0.011281318332897734,
67
+ "acc_norm": 0.6202393906420022,
68
+ "acc_norm_stderr": 0.01132348350471584
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_4.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.338,
5
+ "acc_stderr": 0.014965960710224482
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.346,
9
+ "acc_stderr": 0.015050266127564446
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.33916666666666667,
13
+ "acc_stderr": 0.013672343491681817
14
+ },
15
+ "cb": {
16
+ "acc": 0.5178571428571429,
17
+ "acc_stderr": 0.06737697508644645,
18
+ "f1": 0.3362023995826813
19
+ },
20
+ "copa": {
21
+ "acc": 0.61,
22
+ "acc_stderr": 0.04902071300001975
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2954590718980283,
26
+ "acc_stderr": 0.004553164013379555,
27
+ "acc_norm": 0.32443736307508464,
28
+ "acc_norm_stderr": 0.004672074496749016
29
+ },
30
+ "rte": {
31
+ "acc": 0.48375451263537905,
32
+ "acc_stderr": 0.030080573208738064
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.4846093133385951,
36
+ "acc_stderr": 0.014045826789783656
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5734901122394441,
40
+ "acc_stderr": 0.011436857656268697
41
+ },
42
+ "boolq": {
43
+ "acc": 0.44587155963302755,
44
+ "acc_stderr": 0.008693659886486845
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42634680134680136,
48
+ "acc_stderr": 0.010147858603835144,
49
+ "acc_norm": 0.39057239057239057,
50
+ "acc_norm_stderr": 0.010011059112064239
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19539249146757678,
54
+ "acc_stderr": 0.011586907189952911,
55
+ "acc_norm": 0.23037542662116042,
56
+ "acc_norm_stderr": 0.012304928418747611
57
+ },
58
+ "sciq": {
59
+ "acc": 0.718,
60
+ "acc_stderr": 0.014236526215291334,
61
+ "acc_norm": 0.687,
62
+ "acc_norm_stderr": 0.014671272822977883
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6289445048966268,
66
+ "acc_stderr": 0.011271222398600525,
67
+ "acc_norm": 0.6218715995647442,
68
+ "acc_norm_stderr": 0.011313980666854535
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }
evaluation/rankeval_x_denoiser/checkpoints_2b855b55bc4ul2ndfixnew_4_lm-eval_global_step52452_2023-02-09-23-08-31_4shots_backup.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "anli_r1": {
4
+ "acc": 0.338,
5
+ "acc_stderr": 0.014965960710224482
6
+ },
7
+ "anli_r2": {
8
+ "acc": 0.346,
9
+ "acc_stderr": 0.015050266127564446
10
+ },
11
+ "anli_r3": {
12
+ "acc": 0.33916666666666667,
13
+ "acc_stderr": 0.013672343491681817
14
+ },
15
+ "cb": {
16
+ "acc": 0.5178571428571429,
17
+ "acc_stderr": 0.06737697508644645,
18
+ "f1": 0.3362023995826813
19
+ },
20
+ "copa": {
21
+ "acc": 0.61,
22
+ "acc_stderr": 0.04902071300001975
23
+ },
24
+ "hellaswag": {
25
+ "acc": 0.2954590718980283,
26
+ "acc_stderr": 0.004553164013379555,
27
+ "acc_norm": 0.32443736307508464,
28
+ "acc_norm_stderr": 0.004672074496749016
29
+ },
30
+ "rte": {
31
+ "acc": 0.48375451263537905,
32
+ "acc_stderr": 0.030080573208738064
33
+ },
34
+ "winogrande": {
35
+ "acc": 0.4846093133385951,
36
+ "acc_stderr": 0.014045826789783656
37
+ },
38
+ "storycloze_2016": {
39
+ "acc": 0.5734901122394441,
40
+ "acc_stderr": 0.011436857656268697
41
+ },
42
+ "boolq": {
43
+ "acc": 0.44587155963302755,
44
+ "acc_stderr": 0.008693659886486845
45
+ },
46
+ "arc_easy": {
47
+ "acc": 0.42634680134680136,
48
+ "acc_stderr": 0.010147858603835144,
49
+ "acc_norm": 0.39057239057239057,
50
+ "acc_norm_stderr": 0.010011059112064239
51
+ },
52
+ "arc_challenge": {
53
+ "acc": 0.19539249146757678,
54
+ "acc_stderr": 0.011586907189952911,
55
+ "acc_norm": 0.23037542662116042,
56
+ "acc_norm_stderr": 0.012304928418747611
57
+ },
58
+ "sciq": {
59
+ "acc": 0.718,
60
+ "acc_stderr": 0.014236526215291334,
61
+ "acc_norm": 0.687,
62
+ "acc_norm_stderr": 0.014671272822977883
63
+ },
64
+ "piqa": {
65
+ "acc": 0.6289445048966268,
66
+ "acc_stderr": 0.011271222398600525,
67
+ "acc_norm": 0.6218715995647442,
68
+ "acc_norm_stderr": 0.011313980666854535
69
+ }
70
+ },
71
+ "versions": {
72
+ "anli_r1": 0,
73
+ "anli_r2": 0,
74
+ "anli_r3": 0,
75
+ "cb": 1,
76
+ "copa": 0,
77
+ "hellaswag": 0,
78
+ "rte": 0,
79
+ "winogrande": 0,
80
+ "storycloze_2016": 0,
81
+ "boolq": 1,
82
+ "arc_easy": 0,
83
+ "arc_challenge": 0,
84
+ "sciq": 0,
85
+ "piqa": 0
86
+ }
87
+ }