Hugging Face's logo

datablations
/
lm1-2b8-55b-oscar-repetitions

TensorBoard
Model card Files Metrics Community
Muennighoff commited on
Commit
b415581
•
1 Parent(s): b6c69c9

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. 2b855b11boscar/evaluation/2b855b11boscar_0_babi.json +22 -0
  2. 2b855b11boscar/evaluation/2b855b11boscar_1_babi.json +22 -0
  3. 2b855b11boscar/evaluation/2b855b11boscar_2_babi.json +22 -0
  4. 2b855b11boscar/evaluation/2b855b11boscar_3_babi.json +22 -0
  5. 2b855b11boscar/evaluation/2b855b11boscar_4_babi.json +22 -0
  6. 2b855b11boscar/evaluation/2b855b11boscar_5_babi.json +22 -0
  7. 2b855b11boscar/transformers/merges.txt +0 -0
  8. 2b855b11boscar/transformers/tokenizer.json +0 -0
  9. 2b855b11boscar/transformers/vocab.json +3 -0
  10. 2b855b14boscar/evaluation/2b855b14boscar_0_babi.json +22 -0
  11. 2b855b14boscar/evaluation/2b855b14boscar_1_babi.json +22 -0
  12. 2b855b14boscar/evaluation/2b855b14boscar_2_babi.json +22 -0
  13. 2b855b14boscar/evaluation/2b855b14boscar_3_babi.json +22 -0
  14. 2b855b14boscar/evaluation/2b855b14boscar_4_babi.json +22 -0
  15. 2b855b14boscar/evaluation/2b855b14boscar_5_babi.json +22 -0
  16. 2b855b14boscar/transformers/merges.txt +0 -0
  17. 2b855b14boscar/transformers/tokenizer.json +0 -0
  18. 2b855b14boscar/transformers/vocab.json +3 -0
  19. 2b855b18boscar/evaluation/2b855b18boscar_0_babi.json +22 -0
  20. 2b855b18boscar/evaluation/2b855b18boscar_1_babi.json +22 -0
  21. 2b855b18boscar/evaluation/2b855b18boscar_2_babi.json +22 -0
  22. 2b855b18boscar/evaluation/2b855b18boscar_3_babi.json +22 -0
  23. 2b855b18boscar/evaluation/2b855b18boscar_4_babi.json +22 -0
  24. 2b855b18boscar/evaluation/2b855b18boscar_5_babi.json +22 -0
  25. 2b855b18boscar/transformers/merges.txt +0 -0
  26. 2b855b18boscar/transformers/tokenizer.json +0 -0
  27. 2b855b18boscar/transformers/vocab.json +3 -0
  28. 2b855b1b25oscar/evaluation/2b855b1b25oscar_0_babi.json +22 -0
  29. 2b855b1b25oscar/evaluation/2b855b1b25oscar_1_babi.json +22 -0
  30. 2b855b1b25oscar/evaluation/2b855b1b25oscar_2_babi.json +22 -0
  31. 2b855b1b25oscar/evaluation/2b855b1b25oscar_3_babi.json +22 -0
  32. 2b855b1b25oscar/evaluation/2b855b1b25oscar_4_babi.json +22 -0
  33. 2b855b1b25oscar/evaluation/2b855b1b25oscar_5_babi.json +22 -0
  34. 2b855b1b25oscar/transformers/merges.txt +0 -0
  35. 2b855b1b25oscar/transformers/tokenizer.json +0 -0
  36. 2b855b1b25oscar/transformers/vocab.json +3 -0
  37. 2b855b28boscar/evaluation/2b855b28boscar_0_babi.json +22 -0
  38. 2b855b28boscar/evaluation/2b855b28boscar_1_babi.json +22 -0
  39. 2b855b28boscar/evaluation/2b855b28boscar_2_babi.json +22 -0
  40. 2b855b28boscar/evaluation/2b855b28boscar_3_babi.json +22 -0
  41. 2b855b28boscar/evaluation/2b855b28boscar_4_babi.json +22 -0
  42. 2b855b28boscar/evaluation/2b855b28boscar_5_babi.json +22 -0
  43. 2b855b28boscar/transformers/merges.txt +0 -0
  44. 2b855b28boscar/transformers/tokenizer.json +0 -0
  45. 2b855b28boscar/transformers/vocab.json +3 -0
  46. 2b855b4boscar/evaluation/2b855b4boscar_0_babi.json +22 -0
  47. 2b855b4boscar/evaluation/2b855b4boscar_1_babi.json +22 -0
  48. 2b855b4boscar/evaluation/2b855b4boscar_2_babi.json +22 -0
  49. 2b855b4boscar/evaluation/2b855b4boscar_3_babi.json +22 -0
  50. 2b855b4boscar/evaluation/2b855b4boscar_4_babi.json +22 -0
2b855b11boscar/evaluation/2b855b11boscar_0_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.0,
5
+ "em_stderr": 0.0
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/lm1-2b8-55b-oscar-repetitions/2b855b11boscar/transformers",
14
+ "num_fewshot": 0,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b11boscar/evaluation/2b855b11boscar_1_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.08566666666666667,
5
+ "em_stderr": 0.005110577491542745
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/lm1-2b8-55b-oscar-repetitions/2b855b11boscar/transformers",
14
+ "num_fewshot": 1,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b11boscar/evaluation/2b855b11boscar_2_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.15633333333333332,
5
+ "em_stderr": 0.006631671516023715
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/lm1-2b8-55b-oscar-repetitions/2b855b11boscar/transformers",
14
+ "num_fewshot": 2,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b11boscar/evaluation/2b855b11boscar_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.196,
5
+ "em_stderr": 0.007248828537889569
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/lm1-2b8-55b-oscar-repetitions/2b855b11boscar/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
+ }
2b855b11boscar/evaluation/2b855b11boscar_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.23066666666666666,
5
+ "em_stderr": 0.007692392962423994
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/lm1-2b8-55b-oscar-repetitions/2b855b11boscar/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
+ }
2b855b11boscar/evaluation/2b855b11boscar_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.253,
5
+ "em_stderr": 0.007938388123668571
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/lm1-2b8-55b-oscar-repetitions/2b855b11boscar/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
+ }
2b855b11boscar/transformers/merges.txt ADDED
File without changes
2b855b11boscar/transformers/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
2b855b11boscar/transformers/vocab.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:196139668be63f3b5d6574427317ae82f612a97c5d1cdaf36ed2256dbf636783
3
+ size 1042301
2b855b14boscar/evaluation/2b855b14boscar_0_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.0,
5
+ "em_stderr": 0.0
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/lm1-2b8-55b-oscar-repetitions/2b855b14boscar/transformers",
14
+ "num_fewshot": 0,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b14boscar/evaluation/2b855b14boscar_1_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.074,
5
+ "em_stderr": 0.0047800574944117475
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/lm1-2b8-55b-oscar-repetitions/2b855b14boscar/transformers",
14
+ "num_fewshot": 1,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b14boscar/evaluation/2b855b14boscar_2_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.14466666666666667,
5
+ "em_stderr": 0.0064233813019202715
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/lm1-2b8-55b-oscar-repetitions/2b855b14boscar/transformers",
14
+ "num_fewshot": 2,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b14boscar/evaluation/2b855b14boscar_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.19733333333333333,
5
+ "em_stderr": 0.007267409078059811
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/lm1-2b8-55b-oscar-repetitions/2b855b14boscar/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
+ }
2b855b14boscar/evaluation/2b855b14boscar_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.22433333333333333,
5
+ "em_stderr": 0.007617215389261004
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/lm1-2b8-55b-oscar-repetitions/2b855b14boscar/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
+ }
2b855b14boscar/evaluation/2b855b14boscar_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.25,
5
+ "em_stderr": 0.007907012095608136
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/lm1-2b8-55b-oscar-repetitions/2b855b14boscar/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
+ }
2b855b14boscar/transformers/merges.txt ADDED
File without changes
2b855b14boscar/transformers/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
2b855b14boscar/transformers/vocab.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:196139668be63f3b5d6574427317ae82f612a97c5d1cdaf36ed2256dbf636783
3
+ size 1042301
2b855b18boscar/evaluation/2b855b18boscar_0_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.0,
5
+ "em_stderr": 0.0
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/lm1-2b8-55b-oscar-repetitions/2b855b18boscar/transformers",
14
+ "num_fewshot": 0,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b18boscar/evaluation/2b855b18boscar_1_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.07666666666666666,
5
+ "em_stderr": 0.004858411710742634
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/lm1-2b8-55b-oscar-repetitions/2b855b18boscar/transformers",
14
+ "num_fewshot": 1,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b18boscar/evaluation/2b855b18boscar_2_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.143,
5
+ "em_stderr": 0.006392492053303658
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/lm1-2b8-55b-oscar-repetitions/2b855b18boscar/transformers",
14
+ "num_fewshot": 2,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b18boscar/evaluation/2b855b18boscar_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.19166666666666668,
5
+ "em_stderr": 0.0071875405226526305
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/lm1-2b8-55b-oscar-repetitions/2b855b18boscar/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
+ }
2b855b18boscar/evaluation/2b855b18boscar_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.232,
5
+ "em_stderr": 0.007707905288380708
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/lm1-2b8-55b-oscar-repetitions/2b855b18boscar/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
+ }
2b855b18boscar/evaluation/2b855b18boscar_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.24966666666666668,
5
+ "em_stderr": 0.007903494742862768
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/lm1-2b8-55b-oscar-repetitions/2b855b18boscar/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
+ }
2b855b18boscar/transformers/merges.txt ADDED
File without changes
2b855b18boscar/transformers/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
2b855b18boscar/transformers/vocab.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:196139668be63f3b5d6574427317ae82f612a97c5d1cdaf36ed2256dbf636783
3
+ size 1042301
2b855b1b25oscar/evaluation/2b855b1b25oscar_0_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.0,
5
+ "em_stderr": 0.0
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/lm1-2b8-55b-oscar-repetitions/2b855b1b25oscar/transformers",
14
+ "num_fewshot": 0,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b1b25oscar/evaluation/2b855b1b25oscar_1_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.03866666666666667,
5
+ "em_stderr": 0.0035206036511572203
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/lm1-2b8-55b-oscar-repetitions/2b855b1b25oscar/transformers",
14
+ "num_fewshot": 1,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b1b25oscar/evaluation/2b855b1b25oscar_2_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.059,
5
+ "em_stderr": 0.0043026159679960575
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/lm1-2b8-55b-oscar-repetitions/2b855b1b25oscar/transformers",
14
+ "num_fewshot": 2,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b1b25oscar/evaluation/2b855b1b25oscar_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.07966666666666666,
5
+ "em_stderr": 0.004944503244385582
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/lm1-2b8-55b-oscar-repetitions/2b855b1b25oscar/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
+ }
2b855b1b25oscar/evaluation/2b855b1b25oscar_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.09333333333333334,
5
+ "em_stderr": 0.005311950019695495
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/lm1-2b8-55b-oscar-repetitions/2b855b1b25oscar/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
+ }
2b855b1b25oscar/evaluation/2b855b1b25oscar_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.101,
5
+ "em_stderr": 0.00550240179150498
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/lm1-2b8-55b-oscar-repetitions/2b855b1b25oscar/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
+ }
2b855b1b25oscar/transformers/merges.txt ADDED
File without changes
2b855b1b25oscar/transformers/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
2b855b1b25oscar/transformers/vocab.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:196139668be63f3b5d6574427317ae82f612a97c5d1cdaf36ed2256dbf636783
3
+ size 1042301
2b855b28boscar/evaluation/2b855b28boscar_0_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.0,
5
+ "em_stderr": 0.0
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/lm1-2b8-55b-oscar-repetitions/2b855b28boscar/transformers",
14
+ "num_fewshot": 0,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b28boscar/evaluation/2b855b28boscar_1_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.06633333333333333,
5
+ "em_stderr": 0.004544369171967892
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/lm1-2b8-55b-oscar-repetitions/2b855b28boscar/transformers",
14
+ "num_fewshot": 1,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b28boscar/evaluation/2b855b28boscar_2_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.133,
5
+ "em_stderr": 0.006200791611200593
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/lm1-2b8-55b-oscar-repetitions/2b855b28boscar/transformers",
14
+ "num_fewshot": 2,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b28boscar/evaluation/2b855b28boscar_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.18466666666666667,
5
+ "em_stderr": 0.0070855508823755
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/lm1-2b8-55b-oscar-repetitions/2b855b28boscar/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
+ }
2b855b28boscar/evaluation/2b855b28boscar_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.234,
5
+ "em_stderr": 0.0077309716492093795
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/lm1-2b8-55b-oscar-repetitions/2b855b28boscar/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
+ }
2b855b28boscar/evaluation/2b855b28boscar_5_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.25466666666666665,
5
+ "em_stderr": 0.00795560278820909
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/lm1-2b8-55b-oscar-repetitions/2b855b28boscar/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
+ }
2b855b28boscar/transformers/merges.txt ADDED
File without changes
2b855b28boscar/transformers/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
2b855b28boscar/transformers/vocab.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:196139668be63f3b5d6574427317ae82f612a97c5d1cdaf36ed2256dbf636783
3
+ size 1042301
2b855b4boscar/evaluation/2b855b4boscar_0_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.0,
5
+ "em_stderr": 0.0
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/lm1-2b8-55b-oscar-repetitions/2b855b4boscar/transformers",
14
+ "num_fewshot": 0,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b4boscar/evaluation/2b855b4boscar_1_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.073,
5
+ "em_stderr": 0.004750212785462171
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/lm1-2b8-55b-oscar-repetitions/2b855b4boscar/transformers",
14
+ "num_fewshot": 1,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b4boscar/evaluation/2b855b4boscar_2_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.13633333333333333,
5
+ "em_stderr": 0.006265934819714299
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/lm1-2b8-55b-oscar-repetitions/2b855b4boscar/transformers",
14
+ "num_fewshot": 2,
15
+ "batch_size": null,
16
+ "device": null,
17
+ "no_cache": true,
18
+ "limit": 3000,
19
+ "bootstrap_iters": 100000,
20
+ "description_dict": {}
21
+ }
22
+ }
2b855b4boscar/evaluation/2b855b4boscar_3_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.18,
5
+ "em_stderr": 0.007015440504236936
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/lm1-2b8-55b-oscar-repetitions/2b855b4boscar/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
+ }
2b855b4boscar/evaluation/2b855b4boscar_4_babi.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "babi": {
4
+ "em": 0.22,
5
+ "em_stderr": 0.007564328987051198
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/lm1-2b8-55b-oscar-repetitions/2b855b4boscar/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
+ }