Hugging Face's logo

datablations
/
lm1-4b2-84b-c4pyseeds

Model card Files Community
Muennighoff commited on
Commit
6a79b57
•
1 Parent(s): 5a07b9b

Add

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. 4b284b84b10c4pyseed1/evaluation/4b284b84b10c4pyseed1_4_babi.json +22 -0
  2. 4b284b84b10c4pyseed2/evaluation/4b284b84b10c4pyseed2_4_babi.json +22 -0
  3. 4b284b84b10c4pyseed3/evaluation/4b284b84b10c4pyseed3_3_babi.json +22 -0
  4. 4b284b84b10c4pyseed3/evaluation/4b284b84b10c4pyseed3_4_babi.json +22 -0
  5. 4b284b84b10c4pyseed4/evaluation/4b284b84b10c4pyseed4_3_babi.json +22 -0
  6. 4b284b84b20c4pyseed1/evaluation/4b284b84b20c4pyseed1_3_babi.json +22 -0
  7. 4b284b84b20c4pyseed1/evaluation/4b284b84b20c4pyseed1_4_babi.json +22 -0
  8. 4b284b84b20c4pyseed1/evaluation/4b284b84b20c4pyseed1_5_babi.json +22 -0
  9. 4b284b84b20c4pyseed2/evaluation/4b284b84b20c4pyseed2_3_babi.json +22 -0
  10. 4b284b84b20c4pyseed2/evaluation/4b284b84b20c4pyseed2_4_babi.json +22 -0
  11. 4b284b84b20c4pyseed2/evaluation/4b284b84b20c4pyseed2_5_babi.json +22 -0
  12. 4b284b84b20c4pyseed3/evaluation/4b284b84b20c4pyseed3_3_babi.json +22 -0
  13. 4b284b84b20c4pyseed3/evaluation/4b284b84b20c4pyseed3_4_babi.json +22 -0
  14. 4b284b84b20c4pyseed3/evaluation/4b284b84b20c4pyseed3_5_babi.json +22 -0
  15. 4b284b84b20c4pyseed4/evaluation/4b284b84b20c4pyseed4_3_babi.json +22 -0
  16. 4b284b84b20c4pyseed4/evaluation/4b284b84b20c4pyseed4_4_babi.json +22 -0
  17. 4b284b84b20c4pyseed4/evaluation/4b284b84b20c4pyseed4_5_babi.json +22 -0
  18. 4b284b84b30c4pyseed1/evaluation/4b284b84b30c4pyseed1_3_babi.json +22 -0
  19. 4b284b84b30c4pyseed1/evaluation/4b284b84b30c4pyseed1_4_babi.json +22 -0
  20. 4b284b84b30c4pyseed1/evaluation/4b284b84b30c4pyseed1_5_babi.json +22 -0
  21. 4b284b84b30c4pyseed2/evaluation/4b284b84b30c4pyseed2_3_babi.json +22 -0
  22. 4b284b84b30c4pyseed2/evaluation/4b284b84b30c4pyseed2_4_babi.json +22 -0
  23. 4b284b84b30c4pyseed2/evaluation/4b284b84b30c4pyseed2_5_babi.json +22 -0
  24. 4b284b84b30c4pyseed3/evaluation/4b284b84b30c4pyseed3_3_babi.json +22 -0
  25. 4b284b84b30c4pyseed3/evaluation/4b284b84b30c4pyseed3_4_babi.json +22 -0
  26. 4b284b84b30c4pyseed3/evaluation/4b284b84b30c4pyseed3_5_babi.json +22 -0
  27. 4b284b84b30c4pyseed4/evaluation/4b284b84b30c4pyseed4_3_babi.json +22 -0
  28. 4b284b84b30c4pyseed4/evaluation/4b284b84b30c4pyseed4_4_babi.json +22 -0
  29. 4b284b84b30c4pyseed4/evaluation/4b284b84b30c4pyseed4_5_babi.json +22 -0
  30. 4b284b84b40c4pyseed1/evaluation/4b284b84b40c4pyseed1_3_babi.json +22 -0
  31. 4b284b84b40c4pyseed1/evaluation/4b284b84b40c4pyseed1_4_babi.json +22 -0
  32. 4b284b84b40c4pyseed1/evaluation/4b284b84b40c4pyseed1_5_babi.json +22 -0
  33. 4b284b84b40c4pyseed2/evaluation/4b284b84b40c4pyseed2_3_babi.json +22 -0
  34. 4b284b84b40c4pyseed2/evaluation/4b284b84b40c4pyseed2_4_babi.json +22 -0
  35. 4b284b84b40c4pyseed2/evaluation/4b284b84b40c4pyseed2_5_babi.json +22 -0
  36. 4b284b84b40c4pyseed3/evaluation/4b284b84b40c4pyseed3_3_babi.json +22 -0
  37. 4b284b84b40c4pyseed3/evaluation/4b284b84b40c4pyseed3_4_babi.json +22 -0
  38. 4b284b84b40c4pyseed3/evaluation/4b284b84b40c4pyseed3_5_babi.json +22 -0
  39. 4b284b84b40c4pyseed4/evaluation/4b284b84b40c4pyseed4_3_babi.json +22 -0
  40. 4b284b84b40c4pyseed4/evaluation/4b284b84b40c4pyseed4_4_babi.json +22 -0
  41. 4b284b84b40c4pyseed4/evaluation/4b284b84b40c4pyseed4_5_babi.json +22 -0
  42. 4b284b84b70c4pyseed1/evaluation/4b284b84b70c4pyseed1_3_babi.json +22 -0
  43. 4b284b84b70c4pyseed1/evaluation/4b284b84b70c4pyseed1_4_babi.json +22 -0
  44. 4b284b84b70c4pyseed2/evaluation/4b284b84b70c4pyseed2_3_babi.json +22 -0
  45. 4b284b84b70c4pyseed2/evaluation/4b284b84b70c4pyseed2_4_babi.json +22 -0
  46. 4b284b84b70c4pyseed3/evaluation/4b284b84b70c4pyseed3_3_babi.json +22 -0
  47. 4b284b84b70c4pyseed3/evaluation/4b284b84b70c4pyseed3_4_babi.json +22 -0
  48. 4b284b84b70c4pyseed4/evaluation/4b284b84b70c4pyseed4_3_babi.json +22 -0
  49. 4b284b84b70c4pyseed4/evaluation/4b284b84b70c4pyseed4_4_babi.json +22 -0
  50. 4b284b84b90c4pyseed1/evaluation/4b284b84b90c4pyseed1_3_babi.json +22 -0
4b284b84b10c4pyseed1/evaluation/4b284b84b10c4pyseed1_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.3556666666666667,
5
+ "em_stderr": 0.008741554088695737
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b10c4pyseed1/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b10c4pyseed2/evaluation/4b284b84b10c4pyseed2_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.38133333333333336,
5
+ "em_stderr": 0.008869364649389162
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b10c4pyseed2/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b10c4pyseed3/evaluation/4b284b84b10c4pyseed3_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.2836666666666667,
5
+ "em_stderr": 0.008231396089981295
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b10c4pyseed3/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b10c4pyseed3/evaluation/4b284b84b10c4pyseed3_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.324,
5
+ "em_stderr": 0.008545896331848738
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b10c4pyseed3/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b10c4pyseed4/evaluation/4b284b84b10c4pyseed4_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.332,
5
+ "em_stderr": 0.008599417615616617
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b10c4pyseed4/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b20c4pyseed1/evaluation/4b284b84b20c4pyseed1_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.28733333333333333,
5
+ "em_stderr": 0.008263194888149958
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b20c4pyseed1/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b20c4pyseed1/evaluation/4b284b84b20c4pyseed1_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.32066666666666666,
5
+ "em_stderr": 0.008522757631839066
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b20c4pyseed1/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b20c4pyseed1/evaluation/4b284b84b20c4pyseed1_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.342,
5
+ "em_stderr": 0.008662390682568424
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b20c4pyseed1/transformers",
14
+ "num_fewshot": 5,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b20c4pyseed2/evaluation/4b284b84b20c4pyseed2_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.28633333333333333,
5
+ "em_stderr": 0.008254588497452293
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b20c4pyseed2/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b20c4pyseed2/evaluation/4b284b84b20c4pyseed2_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.32466666666666666,
5
+ "em_stderr": 0.008550464561773713
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b20c4pyseed2/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b20c4pyseed2/evaluation/4b284b84b20c4pyseed2_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.3426666666666667,
5
+ "em_stderr": 0.008666435817412554
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b20c4pyseed2/transformers",
14
+ "num_fewshot": 5,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b20c4pyseed3/evaluation/4b284b84b20c4pyseed3_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.30266666666666664,
5
+ "em_stderr": 0.00838907885685551
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b20c4pyseed3/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b20c4pyseed3/evaluation/4b284b84b20c4pyseed3_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.334,
5
+ "em_stderr": 0.008612358808213386
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b20c4pyseed3/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b20c4pyseed3/evaluation/4b284b84b20c4pyseed3_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.35,
5
+ "em_stderr": 0.008709685386961189
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b20c4pyseed3/transformers",
14
+ "num_fewshot": 5,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b20c4pyseed4/evaluation/4b284b84b20c4pyseed4_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.32366666666666666,
5
+ "em_stderr": 0.008543604796151266
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b20c4pyseed4/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b20c4pyseed4/evaluation/4b284b84b20c4pyseed4_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.3423333333333333,
5
+ "em_stderr": 0.00866441562407849
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b20c4pyseed4/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b20c4pyseed4/evaluation/4b284b84b20c4pyseed4_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.3516666666666667,
5
+ "em_stderr": 0.008719198102229292
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b20c4pyseed4/transformers",
14
+ "num_fewshot": 5,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b30c4pyseed1/evaluation/4b284b84b30c4pyseed1_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.32566666666666666,
5
+ "em_stderr": 0.008557279862345774
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b30c4pyseed1/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b30c4pyseed1/evaluation/4b284b84b30c4pyseed1_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.35133333333333333,
5
+ "em_stderr": 0.008717304889830082
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b30c4pyseed1/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b30c4pyseed1/evaluation/4b284b84b30c4pyseed1_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.3596666666666667,
5
+ "em_stderr": 0.008763244262363716
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b30c4pyseed1/transformers",
14
+ "num_fewshot": 5,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b30c4pyseed2/evaluation/4b284b84b30c4pyseed2_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.336,
5
+ "em_stderr": 0.008625125946859932
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b30c4pyseed2/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b30c4pyseed2/evaluation/4b284b84b30c4pyseed2_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.37133333333333335,
5
+ "em_stderr": 0.008822749567620337
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b30c4pyseed2/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b30c4pyseed2/evaluation/4b284b84b30c4pyseed2_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.396,
5
+ "em_stderr": 0.008930542249025198
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b30c4pyseed2/transformers",
14
+ "num_fewshot": 5,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b30c4pyseed3/evaluation/4b284b84b30c4pyseed3_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.30833333333333335,
5
+ "em_stderr": 0.008432773378427409
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b30c4pyseed3/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b30c4pyseed3/evaluation/4b284b84b30c4pyseed3_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.351,
5
+ "em_stderr": 0.008715407015154936
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b30c4pyseed3/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b30c4pyseed3/evaluation/4b284b84b30c4pyseed3_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.378,
5
+ "em_stderr": 0.00885427200344005
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b30c4pyseed3/transformers",
14
+ "num_fewshot": 5,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b30c4pyseed4/evaluation/4b284b84b30c4pyseed4_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.322,
5
+ "em_stderr": 0.008532072750616225
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b30c4pyseed4/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b30c4pyseed4/evaluation/4b284b84b30c4pyseed4_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.3516666666666667,
5
+ "em_stderr": 0.008719198102229292
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b30c4pyseed4/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b30c4pyseed4/evaluation/4b284b84b30c4pyseed4_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.36766666666666664,
5
+ "em_stderr": 0.008804646702971666
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b30c4pyseed4/transformers",
14
+ "num_fewshot": 5,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b40c4pyseed1/evaluation/4b284b84b40c4pyseed1_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.31533333333333335,
5
+ "em_stderr": 0.008484696268242586
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b40c4pyseed1/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b40c4pyseed1/evaluation/4b284b84b40c4pyseed1_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.349,
5
+ "em_stderr": 0.008703921688004762
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b40c4pyseed1/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b40c4pyseed1/evaluation/4b284b84b40c4pyseed1_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.3566666666666667,
5
+ "em_stderr": 0.008747038855984574
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b40c4pyseed1/transformers",
14
+ "num_fewshot": 5,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b40c4pyseed2/evaluation/4b284b84b40c4pyseed2_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.342,
5
+ "em_stderr": 0.008662390682568435
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b40c4pyseed2/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b40c4pyseed2/evaluation/4b284b84b40c4pyseed2_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.369,
5
+ "em_stderr": 0.008811292732995708
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b40c4pyseed2/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b40c4pyseed2/evaluation/4b284b84b40c4pyseed2_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.38333333333333336,
5
+ "em_stderr": 0.008878207616769261
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b40c4pyseed2/transformers",
14
+ "num_fewshot": 5,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b40c4pyseed3/evaluation/4b284b84b40c4pyseed3_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.316,
5
+ "em_stderr": 0.008489524368773704
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b40c4pyseed3/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b40c4pyseed3/evaluation/4b284b84b40c4pyseed3_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.3546666666666667,
5
+ "em_stderr": 0.008736027709097314
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b40c4pyseed3/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b40c4pyseed3/evaluation/4b284b84b40c4pyseed3_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.37466666666666665,
5
+ "em_stderr": 0.008838734426409927
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b40c4pyseed3/transformers",
14
+ "num_fewshot": 5,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b40c4pyseed4/evaluation/4b284b84b40c4pyseed4_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.30233333333333334,
5
+ "em_stderr": 0.00838646174158945
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b40c4pyseed4/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b40c4pyseed4/evaluation/4b284b84b40c4pyseed4_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.347,
5
+ "em_stderr": 0.008692267742434541
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b40c4pyseed4/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b40c4pyseed4/evaluation/4b284b84b40c4pyseed4_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.359,
5
+ "em_stderr": 0.008759675258642732
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b40c4pyseed4/transformers",
14
+ "num_fewshot": 5,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b70c4pyseed1/evaluation/4b284b84b70c4pyseed1_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.20566666666666666,
5
+ "em_stderr": 0.007380658575174445
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b70c4pyseed1/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b70c4pyseed1/evaluation/4b284b84b70c4pyseed1_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.2673333333333333,
5
+ "em_stderr": 0.008081491874094005
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b70c4pyseed1/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b70c4pyseed2/evaluation/4b284b84b70c4pyseed2_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.251,
5
+ "em_stderr": 0.007917526701966784
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b70c4pyseed2/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b70c4pyseed2/evaluation/4b284b84b70c4pyseed2_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.305,
5
+ "em_stderr": 0.00840725246228854
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b70c4pyseed2/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b70c4pyseed3/evaluation/4b284b84b70c4pyseed3_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.2836666666666667,
5
+ "em_stderr": 0.008231396089981302
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b70c4pyseed3/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b70c4pyseed3/evaluation/4b284b84b70c4pyseed3_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.3313333333333333,
5
+ "em_stderr": 0.00859506507051416
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b70c4pyseed3/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b70c4pyseed4/evaluation/4b284b84b70c4pyseed4_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.293,
5
+ "em_stderr": 0.008311038309737876
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b70c4pyseed4/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b70c4pyseed4/evaluation/4b284b84b70c4pyseed4_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.32133333333333336,
5
+ "em_stderr": 0.008527425152646524
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b70c4pyseed4/transformers",
14
+ "num_fewshot": 4,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
4b284b84b90c4pyseed1/evaluation/4b284b84b90c4pyseed1_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.14,
5
+ "em_stderr": 0.0063361435883239545
6
+ }
7
+ },
8
+ "versions": {
9
+ "babi": 0
10
+ },
11
+ "config": {
12
+ "model": "gpt2",
13
+ "model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4pyseeds/4b284b84b90c4pyseed1/transformers",
14
+ "num_fewshot": 3,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }