Hugging Face's logo

Datasets:
deeplumiere
/
EEE_datastore

License:
Dataset card Files
xet
 Community
21
deeplumiere commited on
Commit
0a163e7
·
1 Parent(s): acd78a0

Automated Adapter Data Update (#20)

Browse files

- Automated data update (2026-07-12) (61d2b66ed3317ad8d44e8003b1e3d6c735de52a4)

This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. data/adapter_stats.json +66 -38
  2. data/global-mmlu-lite/alibaba/qwen3-235b-a22b-instruct-2507/9b88ab5b-4963-4ab9-a381-b2ed8d4c17b5.json +522 -0
  3. data/global-mmlu-lite/anthropic/claude-3-5-haiku-20241022/d180f6bf-a455-4741-915f-cd3105b194ac.json +522 -0
  4. data/global-mmlu-lite/anthropic/claude-3-7-sonnet-20250219/2f5ce580-87fa-4e08-85e1-958db01a560b.json +522 -0
  5. data/global-mmlu-lite/anthropic/claude-opus-4-1-20250805/dff6ca0e-03eb-4fb6-8096-5ccfeb5b66ae.json +522 -0
  6. data/global-mmlu-lite/anthropic/claude-opus-4-8-default/e18bd0fb-8f2c-4016-aafb-6ff1e6fc5a55.json +522 -0
  7. data/global-mmlu-lite/anthropic/claude-sonnet-4-20250514/e2bf6d75-370d-47e0-b027-8b1bd88b4190.json +522 -0
  8. data/global-mmlu-lite/cohere/aya-expanse-32b/c8256802-cef0-441a-8188-b4fd9a447893.json +522 -0
  9. data/global-mmlu-lite/cohere/command-a-03-2025/dc57aed9-944f-4d35-b965-9a800afe812a.json +522 -0
  10. data/global-mmlu-lite/deepseek/deepseek-r1-0528/065c6aa2-454e-4ced-acc9-97bd5d492239.json +522 -0
  11. data/global-mmlu-lite/deepseek/deepseek-v3.1/82ae1a6a-0f6b-4cd2-8196-dc422a8aeaac.json +519 -0
  12. data/global-mmlu-lite/google/gemini-2.5-flash-preview-05-20/3355a584-4bcb-449d-ad27-27ab51bb37d8.json +522 -0
  13. data/global-mmlu-lite/google/gemini-2.5-flash/92f2b2f7-24b1-45cc-b607-0f6b68a5958f.json +522 -0
  14. data/global-mmlu-lite/google/gemini-2.5-pro/cf06df61-6d21-4f86-83a1-41eeae596e48.json +522 -0
  15. data/global-mmlu-lite/google/gemini-3-pro-preview/f4639a49-e254-4dc2-906b-d9466457e7c6.json +522 -0
  16. data/global-mmlu-lite/google/gemini-3.5-flash/8a9501d2-68fa-40d0-8684-4a66ab5c5da8.json +522 -0
  17. data/global-mmlu-lite/google/gemma-3-27b-it/827e6301-a7cd-44af-b981-41d13580fc4b.json +522 -0
  18. data/global-mmlu-lite/google/gemma-3-4b-it/3d4a62a8-a402-49a2-bb72-88d7d0eb342b.json +522 -0
  19. data/global-mmlu-lite/google/gemma-4-26b-a4b-it/ab04e613-6e9c-4682-9723-132b46ed0738.json +522 -0
  20. data/global-mmlu-lite/google/gemma-4-31b-it/2dc4844b-2ada-42a7-9d46-197b50359bc7.json +522 -0
  21. data/global-mmlu-lite/ibm/granite-4.0-h-small/ec935f95-9930-4f99-ac8d-03de3c244dda.json +522 -0
  22. data/global-mmlu-lite/mistralai/mistral-medium-3/6a672cfa-d357-45cb-9ab9-16d5346fcab0.json +522 -0
  23. data/global-mmlu-lite/mistralai/mistral-small-2503/653eafbb-5405-4b59-a20b-0adcdfa0c561.json +522 -0
  24. data/global-mmlu-lite/openai/gpt-4.1-2025-04-14/283f51d5-7745-438b-9f70-792740176eba.json +522 -0
  25. data/global-mmlu-lite/openai/gpt-5-2025-08-07/51afecd5-44e4-4056-bdc2-29b29d8a3cb5.json +522 -0
  26. data/global-mmlu-lite/openai/o3-mini-2025-01-31/68ebbee4-6e28-467f-aca7-12dda3321470.json +522 -0
  27. data/global-mmlu-lite/openai/o4-mini-2025-04-16/350a7b7c-fb17-42b6-8c16-1c220f2863c0.json +522 -0
  28. data/global-mmlu-lite/xai/grok-3-mini/a8bb2b5c-5f6f-42db-8c66-f9eb894f99c0.json +522 -0
  29. data/global-mmlu-lite/xai/grok-4-0709/b6665653-b52f-43d5-9d63-c7a574d1ff9c.json +522 -0
  30. data/rewardbench/0-hero/Matter-0.1-7B-DPO-preview/2fbdc84a-0777-4e21-9143-cb2d9c75cc20.json +134 -0
  31. data/rewardbench/0-hero/Matter-0.1-7B-boost-DPO-preview/ca0344bb-1c2b-4796-8b85-d326fc89a2aa.json +134 -0
  32. data/rewardbench/Ahjeong/MMPO_Gemma_7b/66190e14-2862-4f53-8224-f968e4d55732.json +134 -0
  33. data/rewardbench/Ahjeong/MMPO_Gemma_7b_gamma1.1_epoch3/e96f9bba-8efe-457e-a6b1-4f90b8739f17.json +134 -0
  34. data/rewardbench/Anthropic/claude-3-5-sonnet-20240620/25fcb524-3e9b-4d97-b84d-01986b127c6c.json +116 -0
  35. data/rewardbench/Anthropic/claude-3-haiku-20240307/9421228b-a52b-4b36-a28b-2ad49f2dd5cb.json +134 -0
  36. data/rewardbench/Anthropic/claude-3-opus-20240229/6595e39a-1c58-424f-acc5-c6bff657bc39.json +116 -0
  37. data/rewardbench/Anthropic/claude-3-sonnet-20240229/473f6ca5-f833-4cec-8395-88172151dfc5.json +134 -0
  38. data/rewardbench/AtlaAI/Selene-1-Mini-Llama-3.1-8B/d1f55fa2-64ba-481d-9bff-f82a6dd0ca45.json +116 -0
  39. data/rewardbench/AtlaAI/Selene-1/fd470f89-4dc5-4a7c-aa6b-135d05a9aab6.json +116 -0
  40. data/rewardbench/CIR-AMS/BTRM_Qwen2_7b_0613/216e0174-bab4-40ee-9871-1be675d27445.json +152 -0
  41. data/rewardbench/CIR-AMS/BTRM_Qwen2_7b_0613/3c1e3213-e5ac-47b5-b54a-8d1cbab56e01.json +134 -0
  42. data/rewardbench/CohereForAI/c4ai-command-r-plus/1468c630-4d3a-42b4-b215-c6a457e9d452.json +134 -0
  43. data/rewardbench/ContextualAI/LMUnit-llama3.1-70b/712e97f2-b937-44fd-a27e-6f080a3091da.json +152 -0
  44. data/rewardbench/ContextualAI/LMUnit-qwen2.5-72b/9c657c91-2209-44ca-8bc0-48ec8d7e9bba.json +152 -0
  45. data/rewardbench/ContextualAI/archangel_sft-dpo_llama13b/19e04aeb-d489-48b2-a5ae-2c837f23f686.json +134 -0
  46. data/rewardbench/ContextualAI/archangel_sft-dpo_llama30b/4e7dbb9b-bd79-4e48-8b9a-4a19312e526a.json +134 -0
  47. data/rewardbench/ContextualAI/archangel_sft-dpo_llama7b/2ffd133b-b4c2-4005-8166-42d1012b8397.json +134 -0
  48. data/rewardbench/ContextualAI/archangel_sft-dpo_pythia1-4b/4a4b751a-ec7a-4d99-a0c2-f008cfc389a1.json +134 -0
  49. data/rewardbench/ContextualAI/archangel_sft-dpo_pythia12-0b/48a9c2c0-24f9-467d-9e2d-aabc0a8abdd1.json +134 -0
  50. data/rewardbench/ContextualAI/archangel_sft-dpo_pythia2-8b/7594c6a6-43af-49e5-a2b0-f41448c2e273.json +134 -0
data/adapter_stats.json CHANGED
@@ -30,43 +30,71 @@
30
  "last_failed": true
31
  },
32
  "global-mmlu-lite": {
33
- "data_fingerprint": "5ef9a59574c1b93832c2f2242d9b3258ea081dae80fdf6157c416951eaecc313",
34
  "entry_hashes": [
 
35
  "4651ceff3c95784f330c369c71b6e005369f9d1dc716c6e499d336f3d1a85e14",
36
- "2f6ebe713380721bc675985a219b89f3f6e5d3cb1ac0406bb164950d171e49da",
37
- "617541926325b2677abead9ff73c17b3ea8c4033d3c54a9dbacfcdd2a1675881",
38
  "f0ccfb7d0ca71985c8997a8a977b3c9c299627c13107639240350a4715c75a0e",
39
- "9e78a0d77dc1d35918738b48c17165a29d24b90170c0375e636aae29c1b46e08",
40
- "2bb2af5e0b5152318390230587a97fe6ac3ebf447d4fab5784308d5ff0fb0804",
41
- "0899b9c965496891cbb846b65d0f68c152a81c2b96700941557400d344090da7",
42
- "87c2ec8681f72781d56acdd3e0c9fcd6fa84ac462d166d5d78bddf1e4343b590",
43
- "7d6adaf2431ebaf7f05da4619b32ebde41eb026aea16d3bd8aecec98ac464bdc",
44
- "2895034784c6c5642892b7c992c973047f2146097c6a13dc155f292a5d52867e",
45
- "b18f637032ba10884f65a463b7b28860a671187534bd27097fed2c3041990c28",
46
- "a0dae39f0c46d5c2d066e2516193d212c4952ded26e7767521ed53df91dcb5e9",
47
- "0a6682601e86bd48c3248771f21c03db15ca13fca4c2bb4b52d0abb5bb37340e",
48
  "a55466285b5942f464cd00ec1ef0925e2f7b6a561005b2aad88be014ed62dc11",
49
- "f00df60c831026625621911d570f47fc98a42f82d38cb2a0a2a285e8f8ba870b",
50
- "b47cc80c245540e9e01ce8fd850cc8e071ddb4db644671c4a7b23edd487f93ba",
51
  "cf1434a0eba892b55cd79729aa53923057d629e6d1b264117cce480d172c79e8",
52
- "fee4a524d656eb05c8282f5c9762680946785507fa019968843997cafdf367a0",
53
  "e6fc3bcd91191aeebfcfaa4b52eb256c3759c6d06f98674af0fea97dc9fe8ced",
54
- "a757485a658f13df5452e8a5bb2534332039faaf8c5e0f34c7167ab446b94ae7",
55
- "8e82dcad462992c82b749ea0bd9babb8fddd02ed046775bfc1e4d13af0838216",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
56
  "ec4833f773abcb0d0b375aebbb35cb2846595fd5064b69121011748ae847b211",
57
- "aa570ef42f0c5477de874699226ddee44c86dde134f440e2f3ed5201bc3a59f5",
 
 
 
 
58
  "b1cb0e2c400ce58b6c9dbf2eb67e625fa4a838075a78f01eabac686221c0ba32",
59
- "3afa76ca8cf7122cdda457ef422b1df24025b292bf81dd157f091027e369c233",
60
- "a8853193324cd933332f1e4ee589217e6b756dadc364e81b200410544fc571d8",
 
 
 
 
 
61
  "0783abf527a9d3f8165081d8cb2e33758e68b066ba18eca41c7d97ff06ee2ca9",
62
- "58a9227f563922acc76feee11289f1d6a0e7244f4a9f8194f613a0c12b1ac5b1"
 
 
 
 
 
 
63
  ],
64
- "last_check_ts": 1783601731.979473,
65
- "last_data_change_ts": 1783518896.607522,
66
  "last_failed": false,
67
- "last_success_ts": 1783601731.9794726,
68
- "size_mb": 0,
69
- "time_s": 1.4341261386871338
70
  },
71
  "hal": {
72
  "data_fingerprint": "7bb18f848bce3afee9bf1be8268e2dc115892db8dc93bd8d3daa5cbf21b60ead",
@@ -4992,7 +5020,7 @@
4992
  "url_etag": "W/80766304581529293"
4993
  },
4994
  "hle": {
4995
- "data_fingerprint": "2271963061e62faf7ad57d53687bd26acdc0a5c4824f567d0344498524270633",
4996
  "entry_hashes": [
4997
  "368237761d719dda2f682038eac58a17e6150d09e77db1054eab8813ecb5a2fb",
4998
  "b6ca2fc5888e85a520d37ad92a922794cc1d2a4daf6989d8da6ced059b5d41f3",
@@ -5045,12 +5073,12 @@
5045
  "2d14bf88d84c7b222594b1c3d96605420459a412ad57eb4f01506d73db8e195f",
5046
  "651909ce68911a0049d12fdfd510758a1603338dc60f80a967746a2046b37813"
5047
  ],
5048
- "last_check_ts": 1783601731.8805106,
5049
  "last_data_change_ts": 1783518897.3675156,
5050
  "last_failed": false,
5051
- "last_success_ts": 1783601731.8805096,
5052
  "size_mb": 0,
5053
- "time_s": 1.5643982887268066
5054
  },
5055
  "livecodebenchpro": {
5056
  "last_failed": true
@@ -5538,7 +5566,7 @@
5538
  "time_s": 2.330172300338745
5539
  },
5540
  "multi_swe_bench": {
5541
- "data_fingerprint": "89a0196eeabde307083ea8312416acf3aa63dd9ed391ef39dbbb2f0d1371f3a6",
5542
  "entry_hashes": [
5543
  "f753b7c97aba6b2d8dc03cd7bbd02b8fab2f73d65930442bbe51e984368257c7",
5544
  "a39aee5acb478d354c0844ef682430f74f9ad42c902d3fa25bae02cd21e697a9",
@@ -5803,15 +5831,15 @@
5803
  "a401a3c328a28c918f2fa2d30bc331d2cf739ca5580dd662a5a4e2e12d5abc8e",
5804
  "500a51d34b43447f7354d7c055f958fb02e8f0893529031d802f3c65b80aef42"
5805
  ],
5806
- "last_check_ts": 1783601763.9393828,
5807
  "last_data_change_ts": 1783518930.3473694,
5808
  "last_failed": false,
5809
- "last_success_ts": 1783601763.9393826,
5810
  "size_mb": 0,
5811
- "time_s": 33.76141691207886
5812
  },
5813
  "openeval": {
5814
- "data_fingerprint": "4eba60df14d811b62d1f3b2d77cc2b26e2a6773bc051e2603f5ee8cc8f8315e6",
5815
  "entry_hashes": [
5816
  "6266db2b16a676e3a62eec7195be23f576929bafc1dd8471841609c83f9106e7",
5817
  "550063401f9d81b1e42ba0d5bb04456c5872e2c31c58fd164609e92860f7e36c",
@@ -6926,12 +6954,12 @@
6926
  "592d23a34610414f0f540a25831f981a19790f0fe4434bab39715860fde64f42",
6927
  "6f74e246ae81a8de415c23c5798410c22aaadaddbc9483cb8296de7bf18b2b45"
6928
  ],
6929
- "last_check_ts": 1783602516.2646487,
6930
  "last_data_change_ts": 1783519731.9760447,
6931
  "last_failed": false,
6932
- "last_success_ts": 1783602516.2646482,
6933
  "size_mb": 0,
6934
- "time_s": 785.5236105918884
6935
  },
6936
  "rewardbench": {
6937
  "last_failed": true,
 
30
  "last_failed": true
31
  },
32
  "global-mmlu-lite": {
33
+ "data_fingerprint": "728e4a14097f3a2aa6266537bfe36ed95167f86c0efb8753e537ba1982abe49d",
34
  "entry_hashes": [
35
+ "6b6f6f4959ccd5ca65d63f5061c5681c3aa8e1492676a130ff392eaf80814c6e",
36
  "4651ceff3c95784f330c369c71b6e005369f9d1dc716c6e499d336f3d1a85e14",
37
+ "f00df60c831026625621911d570f47fc98a42f82d38cb2a0a2a285e8f8ba870b",
38
+ "8e82dcad462992c82b749ea0bd9babb8fddd02ed046775bfc1e4d13af0838216",
39
  "f0ccfb7d0ca71985c8997a8a977b3c9c299627c13107639240350a4715c75a0e",
40
+ "13b79500d0bcf8426174175d06a892189da9cf587a3884d253ca3edc6b611150",
41
+ "aa570ef42f0c5477de874699226ddee44c86dde134f440e2f3ed5201bc3a59f5",
 
 
 
 
 
 
 
42
  "a55466285b5942f464cd00ec1ef0925e2f7b6a561005b2aad88be014ed62dc11",
43
+ "2895034784c6c5642892b7c992c973047f2146097c6a13dc155f292a5d52867e",
 
44
  "cf1434a0eba892b55cd79729aa53923057d629e6d1b264117cce480d172c79e8",
 
45
  "e6fc3bcd91191aeebfcfaa4b52eb256c3759c6d06f98674af0fea97dc9fe8ced",
46
+ "97a83d95e9f5ed82780c3492a02666d8b05c70d7d1641aeb0a750bcd91cd82c5",
47
+ "e1654549991d6bd2be1243a7f9d4742c24b80cc6155b78c87b44034989804711",
48
+ "87c2ec8681f72781d56acdd3e0c9fcd6fa84ac462d166d5d78bddf1e4343b590",
49
+ "fee4a524d656eb05c8282f5c9762680946785507fa019968843997cafdf367a0",
50
+ "58a9227f563922acc76feee11289f1d6a0e7244f4a9f8194f613a0c12b1ac5b1",
51
+ "182d7607b74a3e07eb9237cd73e64bc7e6521a198b1063d1cc532786c801712a",
52
+ "94c6b8ccb15f6bdb2bcc32b2b6ab405b36157d4eff55680248489c082d8be3cc",
53
+ "56baf3d5bb3c762fde458787048f09e461e99ad925828c57a98a32ff9ea064af",
54
+ "3afa76ca8cf7122cdda457ef422b1df24025b292bf81dd157f091027e369c233",
55
+ "b47cc80c245540e9e01ce8fd850cc8e071ddb4db644671c4a7b23edd487f93ba",
56
+ "a0dae39f0c46d5c2d066e2516193d212c4952ded26e7767521ed53df91dcb5e9",
57
+ "7ef02ad4eba043ca76f905b2fd6e0c46502c19a1658224240c60828640ca3a83",
58
+ "617541926325b2677abead9ff73c17b3ea8c4033d3c54a9dbacfcdd2a1675881",
59
+ "95907b7006252515d0333b6819f25a9ca3c1527d8e9d9c79c4f2dfd8dc00cdd8",
60
+ "0899b9c965496891cbb846b65d0f68c152a81c2b96700941557400d344090da7",
61
+ "1f96571a09759746d382363de75db25816df1bcb6cbf3ec552b39854cad0b67f",
62
+ "9bbbcfc11fa3bf13eaed47af2835e72c5a143027da7d564f719eb0952e0008c1",
63
+ "039d51e4e23fe38ea363b9b4dbbb26ae227d59ae3c5cd5d40629ba0d4eceeb17",
64
+ "0a6682601e86bd48c3248771f21c03db15ca13fca4c2bb4b52d0abb5bb37340e",
65
+ "5b4feebced3af0e9d6a28fc1de150b82555fdeeaffbe8285f98d759c5f61dae7",
66
+ "7d6adaf2431ebaf7f05da4619b32ebde41eb026aea16d3bd8aecec98ac464bdc",
67
+ "c46c83cec65ac806a031b9ecf8c647d78f3630cea6472a101e3e22ea4b8b766c",
68
+ "bd5a0e428f30a98e50e4da8b130387c744235458a1d069b609b8e1877048d6f1",
69
  "ec4833f773abcb0d0b375aebbb35cb2846595fd5064b69121011748ae847b211",
70
+ "a7b686a4a8594668bd17e5e9cb52fa7a5c93f930ba4569d696bc71c3d4521ed3",
71
+ "9e78a0d77dc1d35918738b48c17165a29d24b90170c0375e636aae29c1b46e08",
72
+ "023eb75410dc101e5369ec8cf3fdeb12a37b865ca3bacfbfaa319bc4ff537c9d",
73
+ "b18f637032ba10884f65a463b7b28860a671187534bd27097fed2c3041990c28",
74
+ "2f6ebe713380721bc675985a219b89f3f6e5d3cb1ac0406bb164950d171e49da",
75
  "b1cb0e2c400ce58b6c9dbf2eb67e625fa4a838075a78f01eabac686221c0ba32",
76
+ "7267510a3e0c1b005475ca19137c440c1b5e3b4039c8da55457b1cecad45586d",
77
+ "b0a82fa63ae27f8eb6c609bc0aed5ad807dcab35eb09e2468494ed03e137ec0c",
78
+ "b79ec1c965d1b92146318c54e61a642afb7a74e1410ad08c82856a17c738118e",
79
+ "a757485a658f13df5452e8a5bb2534332039faaf8c5e0f34c7167ab446b94ae7",
80
+ "1c0598d106085627d47f071c942f8202b6df32a95ba4ae174bc4e42b083d3ea6",
81
+ "a6aab998f015ea0cacae438b9a78f42d90c8057c39ca28bb09e3cb03d396645e",
82
+ "58210e7d713543a1244c08d778845736a8b36523602b55f0eab5a9b4fbcba402",
83
  "0783abf527a9d3f8165081d8cb2e33758e68b066ba18eca41c7d97ff06ee2ca9",
84
+ "a6568d4775433a5de47b14029fa8ec2301eac9b067fdcf241120a9223d1a42fd",
85
+ "2bb2af5e0b5152318390230587a97fe6ac3ebf447d4fab5784308d5ff0fb0804",
86
+ "d506cb354b00b6039fc00a5343f5b40290ae4793f7f76fcc498afd27894955d0",
87
+ "e9fe43ba51dfdf442d3810e3b8dbe62f6333fc2b6c569dd9aead6f400b7f89b7",
88
+ "a8853193324cd933332f1e4ee589217e6b756dadc364e81b200410544fc571d8",
89
+ "cc11505004f63ce30c457bcf5efef7312cf4a8c8342ab62e4ec1d48c88fefc25",
90
+ "940995c822dcd7d8982bad6fd3e80fdbb65d9fbcb75c5d5b1157f52967cd8c67"
91
  ],
92
+ "last_check_ts": 1783820828.8573594,
93
+ "last_data_change_ts": 1783820828.8573608,
94
  "last_failed": false,
95
+ "last_success_ts": 1783820828.8573592,
96
+ "size_mb": 0.3680877685546875,
97
+ "time_s": 1.479100227355957
98
  },
99
  "hal": {
100
  "data_fingerprint": "7bb18f848bce3afee9bf1be8268e2dc115892db8dc93bd8d3daa5cbf21b60ead",
 
5020
  "url_etag": "W/80766304581529293"
5021
  },
5022
  "hle": {
5023
+ "data_fingerprint": "",
5024
  "entry_hashes": [
5025
  "368237761d719dda2f682038eac58a17e6150d09e77db1054eab8813ecb5a2fb",
5026
  "b6ca2fc5888e85a520d37ad92a922794cc1d2a4daf6989d8da6ced059b5d41f3",
 
5073
  "2d14bf88d84c7b222594b1c3d96605420459a412ad57eb4f01506d73db8e195f",
5074
  "651909ce68911a0049d12fdfd510758a1603338dc60f80a967746a2046b37813"
5075
  ],
5076
+ "last_check_ts": 1783820828.8215966,
5077
  "last_data_change_ts": 1783518897.3675156,
5078
  "last_failed": false,
5079
+ "last_success_ts": 1783820828.821596,
5080
  "size_mb": 0,
5081
+ "time_s": 1.448526382446289
5082
  },
5083
  "livecodebenchpro": {
5084
  "last_failed": true
 
5566
  "time_s": 2.330172300338745
5567
  },
5568
  "multi_swe_bench": {
5569
+ "data_fingerprint": "",
5570
  "entry_hashes": [
5571
  "f753b7c97aba6b2d8dc03cd7bbd02b8fab2f73d65930442bbe51e984368257c7",
5572
  "a39aee5acb478d354c0844ef682430f74f9ad42c902d3fa25bae02cd21e697a9",
 
5831
  "a401a3c328a28c918f2fa2d30bc331d2cf739ca5580dd662a5a4e2e12d5abc8e",
5832
  "500a51d34b43447f7354d7c055f958fb02e8f0893529031d802f3c65b80aef42"
5833
  ],
5834
+ "last_check_ts": 1783820857.880977,
5835
  "last_data_change_ts": 1783518930.3473694,
5836
  "last_failed": false,
5837
+ "last_success_ts": 1783820857.8809764,
5838
  "size_mb": 0,
5839
+ "time_s": 30.83329677581787
5840
  },
5841
  "openeval": {
5842
+ "data_fingerprint": "",
5843
  "entry_hashes": [
5844
  "6266db2b16a676e3a62eec7195be23f576929bafc1dd8471841609c83f9106e7",
5845
  "550063401f9d81b1e42ba0d5bb04456c5872e2c31c58fd164609e92860f7e36c",
 
6954
  "592d23a34610414f0f540a25831f981a19790f0fe4434bab39715860fde64f42",
6955
  "6f74e246ae81a8de415c23c5798410c22aaadaddbc9483cb8296de7bf18b2b45"
6956
  ],
6957
+ "last_check_ts": 1783821575.2712789,
6958
  "last_data_change_ts": 1783519731.9760447,
6959
  "last_failed": false,
6960
+ "last_success_ts": 1783821575.2712786,
6961
  "size_mb": 0,
6962
+ "time_s": 747.5695848464966
6963
  },
6964
  "rewardbench": {
6965
  "last_failed": true,
data/global-mmlu-lite/alibaba/qwen3-235b-a22b-instruct-2507/9b88ab5b-4963-4ab9-a381-b2ed8d4c17b5.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/alibaba_qwen3-235b-a22b-instruct-2507/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "qwen3-235b-a22b-instruct-2507",
21
+ "id": "alibaba/qwen3-235b-a22b-instruct-2507",
22
+ "developer": "alibaba",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Qwen 3 235B A22B Instruct 2506"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.8798
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.8522
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.9075
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.88,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0318,
110
+ "upper": 0.0318,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.89,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0307,
137
+ "upper": 0.0307,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.8875,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.031,
164
+ "upper": 0.031,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.885,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0313,
191
+ "upper": 0.0313,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.88,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0318,
218
+ "upper": 0.0318,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.8775,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0321,
245
+ "upper": 0.0321,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.88,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0318,
272
+ "upper": 0.0318,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.88,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0318,
299
+ "upper": 0.0318,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.88,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0318,
326
+ "upper": 0.0318,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.875,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0324,
353
+ "upper": 0.0324,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.8875,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.031,
380
+ "upper": 0.031,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.875,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0324,
407
+ "upper": 0.0324,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.87,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.033,
434
+ "upper": 0.033,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.8725,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0327,
461
+ "upper": 0.0327,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.8775,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0321,
488
+ "upper": 0.0321,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.88,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0318,
515
+ "upper": 0.0318,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/anthropic/claude-3-5-haiku-20241022/d180f6bf-a455-4741-915f-cd3105b194ac.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/anthropic_claude-3-5-haiku-20241022/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "claude-3-5-haiku-20241022",
21
+ "id": "anthropic/claude-3-5-haiku-20241022",
22
+ "developer": "anthropic",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Claude 3.5 Haiku"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.6114
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.5834
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.6394
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.695,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0451,
110
+ "upper": 0.0451,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.485,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.049,
137
+ "upper": 0.049,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.675,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0459,
164
+ "upper": 0.0459,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.565,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0486,
191
+ "upper": 0.0486,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.61,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0478,
218
+ "upper": 0.0478,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.6575,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0465,
245
+ "upper": 0.0465,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.5475,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0488,
272
+ "upper": 0.0488,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.48,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.049,
299
+ "upper": 0.049,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.655,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0466,
326
+ "upper": 0.0466,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.6575,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0465,
353
+ "upper": 0.0465,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.5225,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0489,
380
+ "upper": 0.0489,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.485,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.049,
407
+ "upper": 0.049,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.69,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0453,
434
+ "upper": 0.0453,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.6675,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0462,
461
+ "upper": 0.0462,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.69,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0453,
488
+ "upper": 0.0453,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.7,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0449,
515
+ "upper": 0.0449,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/anthropic/claude-3-7-sonnet-20250219/2f5ce580-87fa-4e08-85e1-958db01a560b.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/anthropic_claude-3-7-sonnet-20250219/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "claude-3-7-sonnet-20250219",
21
+ "id": "anthropic/claude-3-7-sonnet-20250219",
22
+ "developer": "anthropic",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Claude 3.7 Sonnet"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.8078
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.7794
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.8362
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.7925,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0397,
110
+ "upper": 0.0397,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.7625,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0417,
137
+ "upper": 0.0417,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.825,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0372,
164
+ "upper": 0.0372,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.8125,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0382,
191
+ "upper": 0.0382,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.7675,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0414,
218
+ "upper": 0.0414,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.805,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0388,
245
+ "upper": 0.0388,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.8175,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0379,
272
+ "upper": 0.0379,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.8225,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0374,
299
+ "upper": 0.0374,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.8425,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0357,
326
+ "upper": 0.0357,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.83,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0368,
353
+ "upper": 0.0368,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.77,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0412,
380
+ "upper": 0.0412,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.8075,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0386,
407
+ "upper": 0.0386,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.8125,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0382,
434
+ "upper": 0.0382,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.81,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0384,
461
+ "upper": 0.0384,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.835,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0364,
488
+ "upper": 0.0364,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.8125,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0382,
515
+ "upper": 0.0382,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/anthropic/claude-opus-4-1-20250805/dff6ca0e-03eb-4fb6-8096-5ccfeb5b66ae.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/anthropic_claude-opus-4-1-20250805/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "claude-opus-4-1-20250805",
21
+ "id": "anthropic/claude-opus-4-1-20250805",
22
+ "developer": "anthropic",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Claude Opus 4.1"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.943
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.9331
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.9528
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.945,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0223,
110
+ "upper": 0.0223,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.9475,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0219,
137
+ "upper": 0.0219,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.9425,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0228,
164
+ "upper": 0.0228,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.94,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0233,
191
+ "upper": 0.0233,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.945,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0223,
218
+ "upper": 0.0223,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.9475,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0219,
245
+ "upper": 0.0219,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.9425,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0228,
272
+ "upper": 0.0228,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.94,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0233,
299
+ "upper": 0.0233,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.94,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0233,
326
+ "upper": 0.0233,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.95,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0214,
353
+ "upper": 0.0214,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.945,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0223,
380
+ "upper": 0.0223,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.945,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0223,
407
+ "upper": 0.0223,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.93,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.025,
434
+ "upper": 0.025,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.9375,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0237,
461
+ "upper": 0.0237,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.945,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0223,
488
+ "upper": 0.0223,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.945,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0223,
515
+ "upper": 0.0223,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/anthropic/claude-opus-4-8-default/e18bd0fb-8f2c-4016-aafb-6ff1e6fc5a55.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/anthropic_claude-opus-4-8-default/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "claude-opus-4-8-default",
21
+ "id": "anthropic/claude-opus-4-8-default",
22
+ "developer": "anthropic",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Claude Opus 4.8"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.9485
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.9387
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.9584
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.9573,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0199,
110
+ "upper": 0.0199,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.945,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0223,
137
+ "upper": 0.0223,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.9575,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0198,
164
+ "upper": 0.0198,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.95,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0214,
191
+ "upper": 0.0214,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.9425,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0228,
218
+ "upper": 0.0228,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.95,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0214,
245
+ "upper": 0.0214,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.95,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0214,
272
+ "upper": 0.0214,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.9525,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0208,
299
+ "upper": 0.0208,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.9475,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0219,
326
+ "upper": 0.0219,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.945,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0223,
353
+ "upper": 0.0223,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.95,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0214,
380
+ "upper": 0.0214,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.945,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0223,
407
+ "upper": 0.0223,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.945,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0223,
434
+ "upper": 0.0223,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.9494,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0216,
461
+ "upper": 0.0216,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.945,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0223,
488
+ "upper": 0.0223,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.945,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0223,
515
+ "upper": 0.0223,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/anthropic/claude-sonnet-4-20250514/e2bf6d75-370d-47e0-b027-8b1bd88b4190.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/anthropic_claude-sonnet-4-20250514/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "claude-sonnet-4-20250514",
21
+ "id": "anthropic/claude-sonnet-4-20250514",
22
+ "developer": "anthropic",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Claude Sonnet 4"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.9058
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.8913
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.9203
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.9125,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0277,
110
+ "upper": 0.0277,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.905,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0287,
137
+ "upper": 0.0287,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.9075,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0284,
164
+ "upper": 0.0284,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.9125,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0277,
191
+ "upper": 0.0277,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.91,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.028,
218
+ "upper": 0.028,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.9,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0294,
245
+ "upper": 0.0294,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.9025,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0291,
272
+ "upper": 0.0291,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.9075,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0284,
299
+ "upper": 0.0284,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.9,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0294,
326
+ "upper": 0.0294,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.9125,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0277,
353
+ "upper": 0.0277,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.91,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.028,
380
+ "upper": 0.028,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.9075,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0284,
407
+ "upper": 0.0284,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.8975,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0297,
434
+ "upper": 0.0297,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.8975,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0297,
461
+ "upper": 0.0297,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.9175,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.027,
488
+ "upper": 0.027,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.8925,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0304,
515
+ "upper": 0.0304,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/cohere/aya-expanse-32b/c8256802-cef0-441a-8188-b4fd9a447893.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/cohere_aya-expanse-32b/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "aya-expanse-32b",
21
+ "id": "cohere/aya-expanse-32b",
22
+ "developer": "cohere",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Aya Expanse 32B"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.7353
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.6891
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.7815
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.7425,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0429,
110
+ "upper": 0.0429,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.7544,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0422,
137
+ "upper": 0.0422,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.7343,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0433,
164
+ "upper": 0.0433,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.7425,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0429,
191
+ "upper": 0.0429,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.7325,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0434,
218
+ "upper": 0.0434,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.7375,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0431,
245
+ "upper": 0.0431,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.7594,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0419,
272
+ "upper": 0.0419,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.7305,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0436,
299
+ "upper": 0.0436,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.7419,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0429,
326
+ "upper": 0.0429,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.7525,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0423,
353
+ "upper": 0.0423,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.7544,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0422,
380
+ "upper": 0.0422,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.7362,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0433,
407
+ "upper": 0.0433,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.7071,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0448,
434
+ "upper": 0.0448,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.6942,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0452,
461
+ "upper": 0.0452,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.743,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0432,
488
+ "upper": 0.0432,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.7025,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0448,
515
+ "upper": 0.0448,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/cohere/command-a-03-2025/dc57aed9-944f-4d35-b965-9a800afe812a.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/cohere_command-a-03-2025/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "command-a-03-2025",
21
+ "id": "cohere/command-a-03-2025",
22
+ "developer": "cohere",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Command A "
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.8385
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.7993
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.8778
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.8425,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0357,
110
+ "upper": 0.0357,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.855,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0345,
137
+ "upper": 0.0345,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.8225,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0374,
164
+ "upper": 0.0374,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.8425,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0357,
191
+ "upper": 0.0357,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.8375,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0362,
218
+ "upper": 0.0362,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.8421,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0358,
245
+ "upper": 0.0358,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.8546,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0346,
272
+ "upper": 0.0346,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.8375,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0362,
299
+ "upper": 0.0362,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.845,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0355,
326
+ "upper": 0.0355,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.85,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.035,
353
+ "upper": 0.035,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.84,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0359,
380
+ "upper": 0.0359,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.8525,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0348,
407
+ "upper": 0.0348,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.8275,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.037,
434
+ "upper": 0.037,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.815,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0381,
461
+ "upper": 0.0381,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.835,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0364,
488
+ "upper": 0.0364,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.8175,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0379,
515
+ "upper": 0.0379,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/deepseek/deepseek-r1-0528/065c6aa2-454e-4ced-acc9-97bd5d492239.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/deepseek_deepseek-r1-0528/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "deepseek-r1-0528",
21
+ "id": "deepseek/deepseek-r1-0528",
22
+ "developer": "deepseek",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "DeepSeek-R1"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.6744
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.6672
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.6816
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.6825,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0456,
110
+ "upper": 0.0456,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.715,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0442,
137
+ "upper": 0.0442,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.655,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0466,
164
+ "upper": 0.0466,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.6375,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0471,
191
+ "upper": 0.0471,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.6925,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0452,
218
+ "upper": 0.0452,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.6475,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0468,
245
+ "upper": 0.0468,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.655,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0466,
272
+ "upper": 0.0466,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.6775,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0458,
299
+ "upper": 0.0458,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.7725,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0411,
326
+ "upper": 0.0411,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.6575,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0465,
353
+ "upper": 0.0465,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.635,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0472,
380
+ "upper": 0.0472,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.7175,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0441,
407
+ "upper": 0.0441,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.6775,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0458,
434
+ "upper": 0.0458,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.77,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0412,
461
+ "upper": 0.0412,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.5075,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.049,
488
+ "upper": 0.049,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.69,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0453,
515
+ "upper": 0.0453,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/deepseek/deepseek-v3.1/82ae1a6a-0f6b-4cd2-8196-dc422a8aeaac.json ADDED
@@ -0,0 +1,519 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/deepseek_deepseek-v3.1/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "deepseek-v3.1",
21
+ "id": "deepseek/deepseek-v3.1",
22
+ "developer": "deepseek",
23
+ "inference_platform": "unknown"
24
+ },
25
+ "evaluation_results": [
26
+ {
27
+ "evaluation_name": "Average",
28
+ "source_data": {
29
+ "dataset_name": "global-mmlu-lite",
30
+ "source_type": "url",
31
+ "url": [
32
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
33
+ ]
34
+ },
35
+ "metric_config": {
36
+ "evaluation_description": "Global MMLU Lite - Average",
37
+ "lower_is_better": false,
38
+ "score_type": "continuous",
39
+ "min_score": 0.0,
40
+ "max_score": 1.0
41
+ },
42
+ "score_details": {
43
+ "score": 0.8044
44
+ }
45
+ },
46
+ {
47
+ "evaluation_name": "Culturally Sensitive",
48
+ "source_data": {
49
+ "dataset_name": "global-mmlu-lite",
50
+ "source_type": "url",
51
+ "url": [
52
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
53
+ ]
54
+ },
55
+ "metric_config": {
56
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
57
+ "lower_is_better": false,
58
+ "score_type": "continuous",
59
+ "min_score": 0.0,
60
+ "max_score": 1.0
61
+ },
62
+ "score_details": {
63
+ "score": 0.7793
64
+ }
65
+ },
66
+ {
67
+ "evaluation_name": "Culturally Agnostic",
68
+ "source_data": {
69
+ "dataset_name": "global-mmlu-lite",
70
+ "source_type": "url",
71
+ "url": [
72
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
73
+ ]
74
+ },
75
+ "metric_config": {
76
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
77
+ "lower_is_better": false,
78
+ "score_type": "continuous",
79
+ "min_score": 0.0,
80
+ "max_score": 1.0
81
+ },
82
+ "score_details": {
83
+ "score": 0.8295
84
+ }
85
+ },
86
+ {
87
+ "evaluation_name": "Arabic",
88
+ "source_data": {
89
+ "dataset_name": "global-mmlu-lite",
90
+ "source_type": "url",
91
+ "url": [
92
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
93
+ ]
94
+ },
95
+ "metric_config": {
96
+ "evaluation_description": "Global MMLU Lite - Arabic",
97
+ "lower_is_better": false,
98
+ "score_type": "continuous",
99
+ "min_score": 0.0,
100
+ "max_score": 1.0
101
+ },
102
+ "score_details": {
103
+ "score": 0.805,
104
+ "uncertainty": {
105
+ "confidence_interval": {
106
+ "lower": -0.0388,
107
+ "upper": 0.0388,
108
+ "method": "unknown"
109
+ }
110
+ }
111
+ }
112
+ },
113
+ {
114
+ "evaluation_name": "English",
115
+ "source_data": {
116
+ "dataset_name": "global-mmlu-lite",
117
+ "source_type": "url",
118
+ "url": [
119
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
120
+ ]
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Global MMLU Lite - English",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.825,
131
+ "uncertainty": {
132
+ "confidence_interval": {
133
+ "lower": -0.0372,
134
+ "upper": 0.0372,
135
+ "method": "unknown"
136
+ }
137
+ }
138
+ }
139
+ },
140
+ {
141
+ "evaluation_name": "Bengali",
142
+ "source_data": {
143
+ "dataset_name": "global-mmlu-lite",
144
+ "source_type": "url",
145
+ "url": [
146
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
147
+ ]
148
+ },
149
+ "metric_config": {
150
+ "evaluation_description": "Global MMLU Lite - Bengali",
151
+ "lower_is_better": false,
152
+ "score_type": "continuous",
153
+ "min_score": 0.0,
154
+ "max_score": 1.0
155
+ },
156
+ "score_details": {
157
+ "score": 0.8157,
158
+ "uncertainty": {
159
+ "confidence_interval": {
160
+ "lower": -0.0382,
161
+ "upper": 0.0382,
162
+ "method": "unknown"
163
+ }
164
+ }
165
+ }
166
+ },
167
+ {
168
+ "evaluation_name": "German",
169
+ "source_data": {
170
+ "dataset_name": "global-mmlu-lite",
171
+ "source_type": "url",
172
+ "url": [
173
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
174
+ ]
175
+ },
176
+ "metric_config": {
177
+ "evaluation_description": "Global MMLU Lite - German",
178
+ "lower_is_better": false,
179
+ "score_type": "continuous",
180
+ "min_score": 0.0,
181
+ "max_score": 1.0
182
+ },
183
+ "score_details": {
184
+ "score": 0.7925,
185
+ "uncertainty": {
186
+ "confidence_interval": {
187
+ "lower": -0.0397,
188
+ "upper": 0.0397,
189
+ "method": "unknown"
190
+ }
191
+ }
192
+ }
193
+ },
194
+ {
195
+ "evaluation_name": "French",
196
+ "source_data": {
197
+ "dataset_name": "global-mmlu-lite",
198
+ "source_type": "url",
199
+ "url": [
200
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
201
+ ]
202
+ },
203
+ "metric_config": {
204
+ "evaluation_description": "Global MMLU Lite - French",
205
+ "lower_is_better": false,
206
+ "score_type": "continuous",
207
+ "min_score": 0.0,
208
+ "max_score": 1.0
209
+ },
210
+ "score_details": {
211
+ "score": 0.8175,
212
+ "uncertainty": {
213
+ "confidence_interval": {
214
+ "lower": -0.0379,
215
+ "upper": 0.0379,
216
+ "method": "unknown"
217
+ }
218
+ }
219
+ }
220
+ },
221
+ {
222
+ "evaluation_name": "Hindi",
223
+ "source_data": {
224
+ "dataset_name": "global-mmlu-lite",
225
+ "source_type": "url",
226
+ "url": [
227
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
228
+ ]
229
+ },
230
+ "metric_config": {
231
+ "evaluation_description": "Global MMLU Lite - Hindi",
232
+ "lower_is_better": false,
233
+ "score_type": "continuous",
234
+ "min_score": 0.0,
235
+ "max_score": 1.0
236
+ },
237
+ "score_details": {
238
+ "score": 0.7569,
239
+ "uncertainty": {
240
+ "confidence_interval": {
241
+ "lower": -0.0421,
242
+ "upper": 0.0421,
243
+ "method": "unknown"
244
+ }
245
+ }
246
+ }
247
+ },
248
+ {
249
+ "evaluation_name": "Indonesian",
250
+ "source_data": {
251
+ "dataset_name": "global-mmlu-lite",
252
+ "source_type": "url",
253
+ "url": [
254
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
255
+ ]
256
+ },
257
+ "metric_config": {
258
+ "evaluation_description": "Global MMLU Lite - Indonesian",
259
+ "lower_is_better": false,
260
+ "score_type": "continuous",
261
+ "min_score": 0.0,
262
+ "max_score": 1.0
263
+ },
264
+ "score_details": {
265
+ "score": 0.7764,
266
+ "uncertainty": {
267
+ "confidence_interval": {
268
+ "lower": -0.0409,
269
+ "upper": 0.0409,
270
+ "method": "unknown"
271
+ }
272
+ }
273
+ }
274
+ },
275
+ {
276
+ "evaluation_name": "Italian",
277
+ "source_data": {
278
+ "dataset_name": "global-mmlu-lite",
279
+ "source_type": "url",
280
+ "url": [
281
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
282
+ ]
283
+ },
284
+ "metric_config": {
285
+ "evaluation_description": "Global MMLU Lite - Italian",
286
+ "lower_is_better": false,
287
+ "score_type": "continuous",
288
+ "min_score": 0.0,
289
+ "max_score": 1.0
290
+ },
291
+ "score_details": {
292
+ "score": 0.8075,
293
+ "uncertainty": {
294
+ "confidence_interval": {
295
+ "lower": -0.0386,
296
+ "upper": 0.0386,
297
+ "method": "unknown"
298
+ }
299
+ }
300
+ }
301
+ },
302
+ {
303
+ "evaluation_name": "Japanese",
304
+ "source_data": {
305
+ "dataset_name": "global-mmlu-lite",
306
+ "source_type": "url",
307
+ "url": [
308
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
309
+ ]
310
+ },
311
+ "metric_config": {
312
+ "evaluation_description": "Global MMLU Lite - Japanese",
313
+ "lower_is_better": false,
314
+ "score_type": "continuous",
315
+ "min_score": 0.0,
316
+ "max_score": 1.0
317
+ },
318
+ "score_details": {
319
+ "score": 0.8312,
320
+ "uncertainty": {
321
+ "confidence_interval": {
322
+ "lower": -0.0374,
323
+ "upper": 0.0374,
324
+ "method": "unknown"
325
+ }
326
+ }
327
+ }
328
+ },
329
+ {
330
+ "evaluation_name": "Korean",
331
+ "source_data": {
332
+ "dataset_name": "global-mmlu-lite",
333
+ "source_type": "url",
334
+ "url": [
335
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
336
+ ]
337
+ },
338
+ "metric_config": {
339
+ "evaluation_description": "Global MMLU Lite - Korean",
340
+ "lower_is_better": false,
341
+ "score_type": "continuous",
342
+ "min_score": 0.0,
343
+ "max_score": 1.0
344
+ },
345
+ "score_details": {
346
+ "score": 0.8125,
347
+ "uncertainty": {
348
+ "confidence_interval": {
349
+ "lower": -0.0382,
350
+ "upper": 0.0382,
351
+ "method": "unknown"
352
+ }
353
+ }
354
+ }
355
+ },
356
+ {
357
+ "evaluation_name": "Portuguese",
358
+ "source_data": {
359
+ "dataset_name": "global-mmlu-lite",
360
+ "source_type": "url",
361
+ "url": [
362
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
363
+ ]
364
+ },
365
+ "metric_config": {
366
+ "evaluation_description": "Global MMLU Lite - Portuguese",
367
+ "lower_is_better": false,
368
+ "score_type": "continuous",
369
+ "min_score": 0.0,
370
+ "max_score": 1.0
371
+ },
372
+ "score_details": {
373
+ "score": 0.8246,
374
+ "uncertainty": {
375
+ "confidence_interval": {
376
+ "lower": -0.0373,
377
+ "upper": 0.0373,
378
+ "method": "unknown"
379
+ }
380
+ }
381
+ }
382
+ },
383
+ {
384
+ "evaluation_name": "Spanish",
385
+ "source_data": {
386
+ "dataset_name": "global-mmlu-lite",
387
+ "source_type": "url",
388
+ "url": [
389
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
390
+ ]
391
+ },
392
+ "metric_config": {
393
+ "evaluation_description": "Global MMLU Lite - Spanish",
394
+ "lower_is_better": false,
395
+ "score_type": "continuous",
396
+ "min_score": 0.0,
397
+ "max_score": 1.0
398
+ },
399
+ "score_details": {
400
+ "score": 0.8125,
401
+ "uncertainty": {
402
+ "confidence_interval": {
403
+ "lower": -0.0382,
404
+ "upper": 0.0382,
405
+ "method": "unknown"
406
+ }
407
+ }
408
+ }
409
+ },
410
+ {
411
+ "evaluation_name": "Swahili",
412
+ "source_data": {
413
+ "dataset_name": "global-mmlu-lite",
414
+ "source_type": "url",
415
+ "url": [
416
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
417
+ ]
418
+ },
419
+ "metric_config": {
420
+ "evaluation_description": "Global MMLU Lite - Swahili",
421
+ "lower_is_better": false,
422
+ "score_type": "continuous",
423
+ "min_score": 0.0,
424
+ "max_score": 1.0
425
+ },
426
+ "score_details": {
427
+ "score": 0.801,
428
+ "uncertainty": {
429
+ "confidence_interval": {
430
+ "lower": -0.0393,
431
+ "upper": 0.0393,
432
+ "method": "unknown"
433
+ }
434
+ }
435
+ }
436
+ },
437
+ {
438
+ "evaluation_name": "Yoruba",
439
+ "source_data": {
440
+ "dataset_name": "global-mmlu-lite",
441
+ "source_type": "url",
442
+ "url": [
443
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
444
+ ]
445
+ },
446
+ "metric_config": {
447
+ "evaluation_description": "Global MMLU Lite - Yoruba",
448
+ "lower_is_better": false,
449
+ "score_type": "continuous",
450
+ "min_score": 0.0,
451
+ "max_score": 1.0
452
+ },
453
+ "score_details": {
454
+ "score": 0.7831,
455
+ "uncertainty": {
456
+ "confidence_interval": {
457
+ "lower": -0.0415,
458
+ "upper": 0.0415,
459
+ "method": "unknown"
460
+ }
461
+ }
462
+ }
463
+ },
464
+ {
465
+ "evaluation_name": "Chinese",
466
+ "source_data": {
467
+ "dataset_name": "global-mmlu-lite",
468
+ "source_type": "url",
469
+ "url": [
470
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
471
+ ]
472
+ },
473
+ "metric_config": {
474
+ "evaluation_description": "Global MMLU Lite - Chinese",
475
+ "lower_is_better": false,
476
+ "score_type": "continuous",
477
+ "min_score": 0.0,
478
+ "max_score": 1.0
479
+ },
480
+ "score_details": {
481
+ "score": 0.8161,
482
+ "uncertainty": {
483
+ "confidence_interval": {
484
+ "lower": -0.0381,
485
+ "upper": 0.0381,
486
+ "method": "unknown"
487
+ }
488
+ }
489
+ }
490
+ },
491
+ {
492
+ "evaluation_name": "Burmese",
493
+ "source_data": {
494
+ "dataset_name": "global-mmlu-lite",
495
+ "source_type": "url",
496
+ "url": [
497
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
498
+ ]
499
+ },
500
+ "metric_config": {
501
+ "evaluation_description": "Global MMLU Lite - Burmese",
502
+ "lower_is_better": false,
503
+ "score_type": "continuous",
504
+ "min_score": 0.0,
505
+ "max_score": 1.0
506
+ },
507
+ "score_details": {
508
+ "score": 0.7925,
509
+ "uncertainty": {
510
+ "confidence_interval": {
511
+ "lower": -0.0397,
512
+ "upper": 0.0397,
513
+ "method": "unknown"
514
+ }
515
+ }
516
+ }
517
+ }
518
+ ]
519
+ }
data/global-mmlu-lite/google/gemini-2.5-flash-preview-05-20/3355a584-4bcb-449d-ad27-27ab51bb37d8.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/google_gemini-2.5-flash-preview-05-20/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "gemini-2.5-flash-preview-05-20",
21
+ "id": "google/gemini-2.5-flash-preview-05-20",
22
+ "developer": "google",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Gemini 2.5 Flash Preview"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.9092
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.8925
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.9259
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.905,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0287,
110
+ "upper": 0.0287,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.9225,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0262,
137
+ "upper": 0.0262,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.91,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.028,
164
+ "upper": 0.028,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.905,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0287,
191
+ "upper": 0.0287,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.925,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0258,
218
+ "upper": 0.0258,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.9125,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0277,
245
+ "upper": 0.0277,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.9075,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0284,
272
+ "upper": 0.0284,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.89,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0307,
299
+ "upper": 0.0307,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.9125,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0277,
326
+ "upper": 0.0277,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.9075,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0284,
353
+ "upper": 0.0284,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.915,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0273,
380
+ "upper": 0.0273,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.915,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0273,
407
+ "upper": 0.0273,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.905,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0287,
434
+ "upper": 0.0287,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.8825,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0316,
461
+ "upper": 0.0316,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.93,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.025,
488
+ "upper": 0.025,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.9025,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0291,
515
+ "upper": 0.0291,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/google/gemini-2.5-flash/92f2b2f7-24b1-45cc-b607-0f6b68a5958f.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/google_gemini-2.5-flash/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "gemini-2.5-flash",
21
+ "id": "google/gemini-2.5-flash",
22
+ "developer": "google",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Gemini 2.5 Flash"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.9145
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.9
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.9291
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.9125,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0277,
110
+ "upper": 0.0277,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.9325,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0246,
137
+ "upper": 0.0246,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.91,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.028,
164
+ "upper": 0.028,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.9025,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0291,
191
+ "upper": 0.0291,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.91,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.028,
218
+ "upper": 0.028,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.925,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0258,
245
+ "upper": 0.0258,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.9075,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0284,
272
+ "upper": 0.0284,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.9225,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0262,
299
+ "upper": 0.0262,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.9125,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0277,
326
+ "upper": 0.0277,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.915,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0273,
353
+ "upper": 0.0273,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.9125,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0277,
380
+ "upper": 0.0277,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.9175,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.027,
407
+ "upper": 0.027,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.915,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0273,
434
+ "upper": 0.0273,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.9075,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0284,
461
+ "upper": 0.0284,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.915,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0273,
488
+ "upper": 0.0273,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.915,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0273,
515
+ "upper": 0.0273,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/google/gemini-2.5-pro/cf06df61-6d21-4f86-83a1-41eeae596e48.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/google_gemini-2.5-pro/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "gemini-2.5-pro",
21
+ "id": "google/gemini-2.5-pro",
22
+ "developer": "google",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Gemini 2.5 Pro"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.9323
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.9241
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.9406
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.9475,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0219,
110
+ "upper": 0.0219,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.9275,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0254,
137
+ "upper": 0.0254,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.9275,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0254,
164
+ "upper": 0.0254,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.93,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.025,
191
+ "upper": 0.025,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.9425,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0228,
218
+ "upper": 0.0228,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.9275,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0254,
245
+ "upper": 0.0254,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.925,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0258,
272
+ "upper": 0.0258,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.935,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0242,
299
+ "upper": 0.0242,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.9375,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0237,
326
+ "upper": 0.0237,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.9275,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0254,
353
+ "upper": 0.0254,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.93,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.025,
380
+ "upper": 0.025,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.94,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0233,
407
+ "upper": 0.0233,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.9375,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0237,
434
+ "upper": 0.0237,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.925,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0258,
461
+ "upper": 0.0258,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.9275,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0254,
488
+ "upper": 0.0254,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.93,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.025,
515
+ "upper": 0.025,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/google/gemini-3-pro-preview/f4639a49-e254-4dc2-906b-d9466457e7c6.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/google_gemini-3-pro-preview/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "gemini-3-pro-preview",
21
+ "id": "google/gemini-3-pro-preview",
22
+ "developer": "google",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Gemini 3 Pro Preview"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.9453
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.9397
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.9509
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.9475,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0219,
110
+ "upper": 0.0219,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.9425,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0228,
137
+ "upper": 0.0228,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.9425,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0228,
164
+ "upper": 0.0228,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.94,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0233,
191
+ "upper": 0.0233,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.9575,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0198,
218
+ "upper": 0.0198,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.9425,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0228,
245
+ "upper": 0.0228,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.955,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0203,
272
+ "upper": 0.0203,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.955,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0203,
299
+ "upper": 0.0203,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.94,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0233,
326
+ "upper": 0.0233,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.94,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0233,
353
+ "upper": 0.0233,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.9425,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0228,
380
+ "upper": 0.0228,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.9475,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0219,
407
+ "upper": 0.0219,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.94,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0233,
434
+ "upper": 0.0233,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.9425,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0228,
461
+ "upper": 0.0228,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.9475,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0219,
488
+ "upper": 0.0219,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.9425,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0228,
515
+ "upper": 0.0228,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/google/gemini-3.5-flash/8a9501d2-68fa-40d0-8684-4a66ab5c5da8.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/google_gemini-3.5-flash/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "gemini-3.5-flash",
21
+ "id": "google/gemini-3.5-flash",
22
+ "developer": "google",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Gemini 3.5 Flash"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.9539
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.9447
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.9631
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.96,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0192,
110
+ "upper": 0.0192,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.96,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0192,
137
+ "upper": 0.0192,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.9475,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0219,
164
+ "upper": 0.0219,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.9575,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0198,
191
+ "upper": 0.0198,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.9575,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0198,
218
+ "upper": 0.0198,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.9425,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0228,
245
+ "upper": 0.0228,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.9599,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0193,
272
+ "upper": 0.0193,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.95,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0214,
299
+ "upper": 0.0214,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.9575,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0198,
326
+ "upper": 0.0198,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.9525,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0208,
353
+ "upper": 0.0208,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.955,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0203,
380
+ "upper": 0.0203,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.95,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0214,
407
+ "upper": 0.0214,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.945,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0223,
434
+ "upper": 0.0223,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.96,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0192,
461
+ "upper": 0.0192,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.9525,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0208,
488
+ "upper": 0.0208,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.955,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0203,
515
+ "upper": 0.0203,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/google/gemma-3-27b-it/827e6301-a7cd-44af-b981-41d13580fc4b.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/google_gemma-3-27b-it/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "gemma-3-27b-it",
21
+ "id": "google/gemma-3-27b-it",
22
+ "developer": "google",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Gemma 3 27B"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.763
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.7528
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.7733
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.78,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0406,
110
+ "upper": 0.0406,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.7337,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0434,
137
+ "upper": 0.0434,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.75,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0426,
164
+ "upper": 0.0426,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.775,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0409,
191
+ "upper": 0.0409,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.7481,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0429,
218
+ "upper": 0.0429,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.7335,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0437,
245
+ "upper": 0.0437,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.7563,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0422,
272
+ "upper": 0.0422,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.75,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0424,
299
+ "upper": 0.0424,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.7925,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0397,
326
+ "upper": 0.0397,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.798,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0395,
353
+ "upper": 0.0395,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.7481,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0427,
380
+ "upper": 0.0427,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.7494,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0425,
407
+ "upper": 0.0425,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.785,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0403,
434
+ "upper": 0.0403,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.7444,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0428,
461
+ "upper": 0.0428,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.7925,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0397,
488
+ "upper": 0.0397,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.7719,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0412,
515
+ "upper": 0.0412,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/google/gemma-3-4b-it/3d4a62a8-a402-49a2-bb72-88d7d0eb342b.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/google_gemma-3-4b-it/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "gemma-3-4b-it",
21
+ "id": "google/gemma-3-4b-it",
22
+ "developer": "google",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Gemma 3 4B"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.6511
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.6116
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.6906
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.6525,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0467,
110
+ "upper": 0.0467,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.67,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0461,
137
+ "upper": 0.0461,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.68,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0457,
164
+ "upper": 0.0457,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.6525,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0467,
191
+ "upper": 0.0467,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.6575,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0465,
218
+ "upper": 0.0465,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.6475,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0468,
245
+ "upper": 0.0468,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.6775,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0458,
272
+ "upper": 0.0458,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.6675,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0462,
299
+ "upper": 0.0462,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.6325,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0472,
326
+ "upper": 0.0472,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.66,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0464,
353
+ "upper": 0.0464,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.68,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0457,
380
+ "upper": 0.0457,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.6725,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.046,
407
+ "upper": 0.046,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.6075,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0479,
434
+ "upper": 0.0479,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.5825,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0483,
461
+ "upper": 0.0483,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.6475,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0468,
488
+ "upper": 0.0468,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.63,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0473,
515
+ "upper": 0.0473,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/google/gemma-4-26b-a4b-it/ab04e613-6e9c-4682-9723-132b46ed0738.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/google_gemma-4-26b-a4b-it/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "gemma-4-26b-a4b-it",
21
+ "id": "google/gemma-4-26b-a4b-it",
22
+ "developer": "google",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Gemma 4 26B A4B"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.8571
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.8465
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.8677
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.8596,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0341,
110
+ "upper": 0.0341,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.8797,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0319,
137
+ "upper": 0.0319,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.8308,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0369,
164
+ "upper": 0.0369,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.8596,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0341,
191
+ "upper": 0.0341,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.8546,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0346,
218
+ "upper": 0.0346,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.8425,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0357,
245
+ "upper": 0.0357,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.8521,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0348,
272
+ "upper": 0.0348,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.8546,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0346,
299
+ "upper": 0.0346,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.8722,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0328,
326
+ "upper": 0.0328,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.8766,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0324,
353
+ "upper": 0.0324,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.8467,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0354,
380
+ "upper": 0.0354,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.8675,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0332,
407
+ "upper": 0.0332,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.855,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0345,
434
+ "upper": 0.0345,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.8246,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0373,
461
+ "upper": 0.0373,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.8797,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0319,
488
+ "upper": 0.0319,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.8571,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0343,
515
+ "upper": 0.0343,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/google/gemma-4-31b-it/2dc4844b-2ada-42a7-9d46-197b50359bc7.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/google_gemma-4-31b-it/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "gemma-4-31b-it",
21
+ "id": "google/gemma-4-31b-it",
22
+ "developer": "google",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Gemma 4 31B"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.9068
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.8925
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.9212
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.9125,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0277,
110
+ "upper": 0.0277,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.905,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0287,
137
+ "upper": 0.0287,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.905,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0287,
164
+ "upper": 0.0287,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.9175,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.027,
191
+ "upper": 0.027,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.91,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.028,
218
+ "upper": 0.028,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.91,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.028,
245
+ "upper": 0.028,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.9075,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0284,
272
+ "upper": 0.0284,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.9025,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0291,
299
+ "upper": 0.0291,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.9025,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0291,
326
+ "upper": 0.0291,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.9,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0294,
353
+ "upper": 0.0294,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.92,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0266,
380
+ "upper": 0.0266,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.9125,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0277,
407
+ "upper": 0.0277,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.902,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0292,
434
+ "upper": 0.0292,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.8975,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0297,
461
+ "upper": 0.0297,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.9075,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0284,
488
+ "upper": 0.0284,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.8975,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0297,
515
+ "upper": 0.0297,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/ibm/granite-4.0-h-small/ec935f95-9930-4f99-ac8d-03de3c244dda.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/ibm_granite-4.0-h-small/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "granite-4.0-h-small",
21
+ "id": "ibm/granite-4.0-h-small",
22
+ "developer": "ibm",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Granite 4.0 Small"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.7503
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.7182
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.7826
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.7613,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0419,
110
+ "upper": 0.0419,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.77,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0412,
137
+ "upper": 0.0412,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.7613,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0419,
164
+ "upper": 0.0419,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.755,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0421,
191
+ "upper": 0.0421,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.7594,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0419,
218
+ "upper": 0.0419,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.7575,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.042,
245
+ "upper": 0.042,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.7614,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0421,
272
+ "upper": 0.0421,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.7525,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0423,
299
+ "upper": 0.0423,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.7406,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0431,
326
+ "upper": 0.0431,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.7525,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0423,
353
+ "upper": 0.0423,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.757,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0423,
380
+ "upper": 0.0423,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.7638,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0417,
407
+ "upper": 0.0417,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.7318,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0435,
434
+ "upper": 0.0435,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.6921,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0456,
461
+ "upper": 0.0456,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.7475,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0426,
488
+ "upper": 0.0426,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.7419,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0429,
515
+ "upper": 0.0429,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/mistralai/mistral-medium-3/6a672cfa-d357-45cb-9ab9-16d5346fcab0.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/mistralai_mistral-medium-3/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "mistral-medium-3",
21
+ "id": "mistralai/mistral-medium-3",
22
+ "developer": "mistralai",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Mistral Medium 3"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.5511
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.5391
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.5631
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.455,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0488,
110
+ "upper": 0.0488,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.38,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0476,
137
+ "upper": 0.0476,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.5175,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.049,
164
+ "upper": 0.049,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.4775,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0489,
191
+ "upper": 0.0489,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.41,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0482,
218
+ "upper": 0.0482,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.555,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0487,
245
+ "upper": 0.0487,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.515,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.049,
272
+ "upper": 0.049,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.535,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0489,
299
+ "upper": 0.0489,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.58,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0484,
326
+ "upper": 0.0484,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.595,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0481,
353
+ "upper": 0.0481,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.5175,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.049,
380
+ "upper": 0.049,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.5375,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0489,
407
+ "upper": 0.0489,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.7075,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0446,
434
+ "upper": 0.0446,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.7675,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0414,
461
+ "upper": 0.0414,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.535,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0489,
488
+ "upper": 0.0489,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.7325,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0434,
515
+ "upper": 0.0434,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/mistralai/mistral-small-2503/653eafbb-5405-4b59-a20b-0adcdfa0c561.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/mistralai_mistral-small-2503/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "mistral-small-2503",
21
+ "id": "mistralai/mistral-small-2503",
22
+ "developer": "mistralai",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Mistral Small 3.1"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.7852
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.7537
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.8166
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.7875,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0401,
110
+ "upper": 0.0401,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.8,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0392,
137
+ "upper": 0.0392,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.7725,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0411,
164
+ "upper": 0.0411,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.7975,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0394,
191
+ "upper": 0.0394,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.8,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0392,
218
+ "upper": 0.0392,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.795,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0396,
245
+ "upper": 0.0396,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.785,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0403,
272
+ "upper": 0.0403,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.805,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0388,
299
+ "upper": 0.0388,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.77,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0412,
326
+ "upper": 0.0412,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.79,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0399,
353
+ "upper": 0.0399,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.7925,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0397,
380
+ "upper": 0.0397,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.7825,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0404,
407
+ "upper": 0.0404,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.775,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0409,
434
+ "upper": 0.0409,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.735,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0432,
461
+ "upper": 0.0432,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.7925,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0397,
488
+ "upper": 0.0397,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.7825,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0404,
515
+ "upper": 0.0404,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/openai/gpt-4.1-2025-04-14/283f51d5-7745-438b-9f70-792740176eba.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/openai_gpt-4.1-2025-04-14/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "gpt-4.1-2025-04-14",
21
+ "id": "openai/gpt-4.1-2025-04-14",
22
+ "developer": "openai",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "GPT-4.1"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.8755
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.8541
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.8969
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.88,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0318,
110
+ "upper": 0.0318,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.8825,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0316,
137
+ "upper": 0.0316,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.8625,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0337,
164
+ "upper": 0.0337,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.875,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0324,
191
+ "upper": 0.0324,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.8875,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.031,
218
+ "upper": 0.031,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.8775,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0321,
245
+ "upper": 0.0321,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.885,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0313,
272
+ "upper": 0.0313,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.88,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0318,
299
+ "upper": 0.0318,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.8725,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0327,
326
+ "upper": 0.0327,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.87,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.033,
353
+ "upper": 0.033,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.875,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0324,
380
+ "upper": 0.0324,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.885,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0313,
407
+ "upper": 0.0313,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.8725,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0327,
434
+ "upper": 0.0327,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.875,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0324,
461
+ "upper": 0.0324,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.87,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.033,
488
+ "upper": 0.033,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.8575,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0343,
515
+ "upper": 0.0343,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/openai/gpt-5-2025-08-07/51afecd5-44e4-4056-bdc2-29b29d8a3cb5.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/openai_gpt-5-2025-08-07/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "gpt-5-2025-08-07",
21
+ "id": "openai/gpt-5-2025-08-07",
22
+ "developer": "openai",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "GPT-5"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.8895
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.8913
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.8878
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.8925,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0304,
110
+ "upper": 0.0304,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.8725,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0327,
137
+ "upper": 0.0327,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.9,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0294,
164
+ "upper": 0.0294,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.91,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.028,
191
+ "upper": 0.028,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.9075,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0284,
218
+ "upper": 0.0284,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.865,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0335,
245
+ "upper": 0.0335,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.795,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0396,
272
+ "upper": 0.0396,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.9075,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0284,
299
+ "upper": 0.0284,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.8875,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.031,
326
+ "upper": 0.031,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.915,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0273,
353
+ "upper": 0.0273,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.8875,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.031,
380
+ "upper": 0.031,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.905,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0287,
407
+ "upper": 0.0287,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.865,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0335,
434
+ "upper": 0.0335,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.9125,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0277,
461
+ "upper": 0.0277,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.895,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.03,
488
+ "upper": 0.03,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.915,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0273,
515
+ "upper": 0.0273,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/openai/o3-mini-2025-01-31/68ebbee4-6e28-467f-aca7-12dda3321470.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/openai_o3-mini-2025-01-31/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "o3-mini-2025-01-31",
21
+ "id": "openai/o3-mini-2025-01-31",
22
+ "developer": "openai",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "o3 mini"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.78
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.765
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.795
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.7725,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0411,
110
+ "upper": 0.0411,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.8025,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.039,
137
+ "upper": 0.039,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.77,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0412,
164
+ "upper": 0.0412,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.7525,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0423,
191
+ "upper": 0.0423,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.74,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.043,
218
+ "upper": 0.043,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.7525,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0423,
245
+ "upper": 0.0423,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.7425,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0429,
272
+ "upper": 0.0429,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.8,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0392,
299
+ "upper": 0.0392,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.81,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0384,
326
+ "upper": 0.0384,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.8075,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0386,
353
+ "upper": 0.0386,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.7975,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0394,
380
+ "upper": 0.0394,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.775,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0409,
407
+ "upper": 0.0409,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.765,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0416,
434
+ "upper": 0.0416,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.7725,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0411,
461
+ "upper": 0.0411,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.8125,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0382,
488
+ "upper": 0.0382,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.8075,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0386,
515
+ "upper": 0.0386,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/openai/o4-mini-2025-04-16/350a7b7c-fb17-42b6-8c16-1c220f2863c0.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/openai_o4-mini-2025-04-16/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "o4-mini-2025-04-16",
21
+ "id": "openai/o4-mini-2025-04-16",
22
+ "developer": "openai",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "o4 mini"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.8705
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.8503
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.8906
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.865,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0335,
110
+ "upper": 0.0335,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.8675,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0332,
137
+ "upper": 0.0332,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.8875,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.031,
164
+ "upper": 0.031,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.8775,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0321,
191
+ "upper": 0.0321,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.87,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.033,
218
+ "upper": 0.033,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.87,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.033,
245
+ "upper": 0.033,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.8675,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0332,
272
+ "upper": 0.0332,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.855,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0345,
299
+ "upper": 0.0345,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.885,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0313,
326
+ "upper": 0.0313,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.88,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0318,
353
+ "upper": 0.0318,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.88,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0318,
380
+ "upper": 0.0318,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.855,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0345,
407
+ "upper": 0.0345,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.8525,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0348,
434
+ "upper": 0.0348,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.8525,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0348,
461
+ "upper": 0.0348,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.89,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0307,
488
+ "upper": 0.0307,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.8725,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0327,
515
+ "upper": 0.0327,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/xai/grok-3-mini/a8bb2b5c-5f6f-42db-8c66-f9eb894f99c0.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/xai_grok-3-mini/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "grok-3-mini",
21
+ "id": "xai/grok-3-mini",
22
+ "developer": "xai",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Grok 3 Mini"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.673
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.6717
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.6743
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.755,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0421,
110
+ "upper": 0.0421,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.5075,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.049,
137
+ "upper": 0.049,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.7355,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0434,
164
+ "upper": 0.0434,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.6591,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0465,
191
+ "upper": 0.0465,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.485,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.049,
218
+ "upper": 0.049,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.56,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0486,
245
+ "upper": 0.0486,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.725,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0438,
272
+ "upper": 0.0438,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.696,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0452,
299
+ "upper": 0.0452,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.6575,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.0465,
326
+ "upper": 0.0465,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.7325,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.0434,
353
+ "upper": 0.0434,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.6275,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0474,
380
+ "upper": 0.0474,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.61,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0478,
407
+ "upper": 0.0478,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.7625,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.0417,
434
+ "upper": 0.0417,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.8296,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0369,
461
+ "upper": 0.0369,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.5564,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0487,
488
+ "upper": 0.0487,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.8693,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0331,
515
+ "upper": 0.0331,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/global-mmlu-lite/xai/grok-4-0709/b6665653-b52f-43d5-9d63-c7a574d1ff9c.json ADDED
@@ -0,0 +1,522 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.2",
3
+ "evaluation_id": "global-mmlu-lite/xai_grok-4-0709/1783820827.6963763",
4
+ "retrieved_timestamp": "1783820827.6963763",
5
+ "source_metadata": {
6
+ "source_name": "Global MMLU Lite Leaderboard",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "kaggle",
9
+ "source_organization_url": "www.kaggle.com",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "kaggle kernel",
14
+ "version": "4",
15
+ "additional_details": {
16
+ "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook"
17
+ }
18
+ },
19
+ "model_info": {
20
+ "name": "grok-4-0709",
21
+ "id": "xai/grok-4-0709",
22
+ "developer": "xai",
23
+ "inference_platform": "unknown",
24
+ "additional_details": {
25
+ "display_name": "Grok 4"
26
+ }
27
+ },
28
+ "evaluation_results": [
29
+ {
30
+ "evaluation_name": "Average",
31
+ "source_data": {
32
+ "dataset_name": "global-mmlu-lite",
33
+ "source_type": "url",
34
+ "url": [
35
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
36
+ ]
37
+ },
38
+ "metric_config": {
39
+ "evaluation_description": "Global MMLU Lite - Average",
40
+ "lower_is_better": false,
41
+ "score_type": "continuous",
42
+ "min_score": 0.0,
43
+ "max_score": 1.0
44
+ },
45
+ "score_details": {
46
+ "score": 0.8881
47
+ }
48
+ },
49
+ {
50
+ "evaluation_name": "Culturally Sensitive",
51
+ "source_data": {
52
+ "dataset_name": "global-mmlu-lite",
53
+ "source_type": "url",
54
+ "url": [
55
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
56
+ ]
57
+ },
58
+ "metric_config": {
59
+ "evaluation_description": "Global MMLU Lite - Culturally Sensitive",
60
+ "lower_is_better": false,
61
+ "score_type": "continuous",
62
+ "min_score": 0.0,
63
+ "max_score": 1.0
64
+ },
65
+ "score_details": {
66
+ "score": 0.8862
67
+ }
68
+ },
69
+ {
70
+ "evaluation_name": "Culturally Agnostic",
71
+ "source_data": {
72
+ "dataset_name": "global-mmlu-lite",
73
+ "source_type": "url",
74
+ "url": [
75
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
76
+ ]
77
+ },
78
+ "metric_config": {
79
+ "evaluation_description": "Global MMLU Lite - Culturally Agnostic",
80
+ "lower_is_better": false,
81
+ "score_type": "continuous",
82
+ "min_score": 0.0,
83
+ "max_score": 1.0
84
+ },
85
+ "score_details": {
86
+ "score": 0.89
87
+ }
88
+ },
89
+ {
90
+ "evaluation_name": "Arabic",
91
+ "source_data": {
92
+ "dataset_name": "global-mmlu-lite",
93
+ "source_type": "url",
94
+ "url": [
95
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
96
+ ]
97
+ },
98
+ "metric_config": {
99
+ "evaluation_description": "Global MMLU Lite - Arabic",
100
+ "lower_is_better": false,
101
+ "score_type": "continuous",
102
+ "min_score": 0.0,
103
+ "max_score": 1.0
104
+ },
105
+ "score_details": {
106
+ "score": 0.885,
107
+ "uncertainty": {
108
+ "confidence_interval": {
109
+ "lower": -0.0313,
110
+ "upper": 0.0313,
111
+ "method": "unknown"
112
+ }
113
+ }
114
+ }
115
+ },
116
+ {
117
+ "evaluation_name": "English",
118
+ "source_data": {
119
+ "dataset_name": "global-mmlu-lite",
120
+ "source_type": "url",
121
+ "url": [
122
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
123
+ ]
124
+ },
125
+ "metric_config": {
126
+ "evaluation_description": "Global MMLU Lite - English",
127
+ "lower_is_better": false,
128
+ "score_type": "continuous",
129
+ "min_score": 0.0,
130
+ "max_score": 1.0
131
+ },
132
+ "score_details": {
133
+ "score": 0.905,
134
+ "uncertainty": {
135
+ "confidence_interval": {
136
+ "lower": -0.0287,
137
+ "upper": 0.0287,
138
+ "method": "unknown"
139
+ }
140
+ }
141
+ }
142
+ },
143
+ {
144
+ "evaluation_name": "Bengali",
145
+ "source_data": {
146
+ "dataset_name": "global-mmlu-lite",
147
+ "source_type": "url",
148
+ "url": [
149
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
150
+ ]
151
+ },
152
+ "metric_config": {
153
+ "evaluation_description": "Global MMLU Lite - Bengali",
154
+ "lower_is_better": false,
155
+ "score_type": "continuous",
156
+ "min_score": 0.0,
157
+ "max_score": 1.0
158
+ },
159
+ "score_details": {
160
+ "score": 0.8925,
161
+ "uncertainty": {
162
+ "confidence_interval": {
163
+ "lower": -0.0304,
164
+ "upper": 0.0304,
165
+ "method": "unknown"
166
+ }
167
+ }
168
+ }
169
+ },
170
+ {
171
+ "evaluation_name": "German",
172
+ "source_data": {
173
+ "dataset_name": "global-mmlu-lite",
174
+ "source_type": "url",
175
+ "url": [
176
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
177
+ ]
178
+ },
179
+ "metric_config": {
180
+ "evaluation_description": "Global MMLU Lite - German",
181
+ "lower_is_better": false,
182
+ "score_type": "continuous",
183
+ "min_score": 0.0,
184
+ "max_score": 1.0
185
+ },
186
+ "score_details": {
187
+ "score": 0.8725,
188
+ "uncertainty": {
189
+ "confidence_interval": {
190
+ "lower": -0.0327,
191
+ "upper": 0.0327,
192
+ "method": "unknown"
193
+ }
194
+ }
195
+ }
196
+ },
197
+ {
198
+ "evaluation_name": "French",
199
+ "source_data": {
200
+ "dataset_name": "global-mmlu-lite",
201
+ "source_type": "url",
202
+ "url": [
203
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
204
+ ]
205
+ },
206
+ "metric_config": {
207
+ "evaluation_description": "Global MMLU Lite - French",
208
+ "lower_is_better": false,
209
+ "score_type": "continuous",
210
+ "min_score": 0.0,
211
+ "max_score": 1.0
212
+ },
213
+ "score_details": {
214
+ "score": 0.875,
215
+ "uncertainty": {
216
+ "confidence_interval": {
217
+ "lower": -0.0324,
218
+ "upper": 0.0324,
219
+ "method": "unknown"
220
+ }
221
+ }
222
+ }
223
+ },
224
+ {
225
+ "evaluation_name": "Hindi",
226
+ "source_data": {
227
+ "dataset_name": "global-mmlu-lite",
228
+ "source_type": "url",
229
+ "url": [
230
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
231
+ ]
232
+ },
233
+ "metric_config": {
234
+ "evaluation_description": "Global MMLU Lite - Hindi",
235
+ "lower_is_better": false,
236
+ "score_type": "continuous",
237
+ "min_score": 0.0,
238
+ "max_score": 1.0
239
+ },
240
+ "score_details": {
241
+ "score": 0.8675,
242
+ "uncertainty": {
243
+ "confidence_interval": {
244
+ "lower": -0.0332,
245
+ "upper": 0.0332,
246
+ "method": "unknown"
247
+ }
248
+ }
249
+ }
250
+ },
251
+ {
252
+ "evaluation_name": "Indonesian",
253
+ "source_data": {
254
+ "dataset_name": "global-mmlu-lite",
255
+ "source_type": "url",
256
+ "url": [
257
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
258
+ ]
259
+ },
260
+ "metric_config": {
261
+ "evaluation_description": "Global MMLU Lite - Indonesian",
262
+ "lower_is_better": false,
263
+ "score_type": "continuous",
264
+ "min_score": 0.0,
265
+ "max_score": 1.0
266
+ },
267
+ "score_details": {
268
+ "score": 0.89,
269
+ "uncertainty": {
270
+ "confidence_interval": {
271
+ "lower": -0.0307,
272
+ "upper": 0.0307,
273
+ "method": "unknown"
274
+ }
275
+ }
276
+ }
277
+ },
278
+ {
279
+ "evaluation_name": "Italian",
280
+ "source_data": {
281
+ "dataset_name": "global-mmlu-lite",
282
+ "source_type": "url",
283
+ "url": [
284
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
285
+ ]
286
+ },
287
+ "metric_config": {
288
+ "evaluation_description": "Global MMLU Lite - Italian",
289
+ "lower_is_better": false,
290
+ "score_type": "continuous",
291
+ "min_score": 0.0,
292
+ "max_score": 1.0
293
+ },
294
+ "score_details": {
295
+ "score": 0.9025,
296
+ "uncertainty": {
297
+ "confidence_interval": {
298
+ "lower": -0.0291,
299
+ "upper": 0.0291,
300
+ "method": "unknown"
301
+ }
302
+ }
303
+ }
304
+ },
305
+ {
306
+ "evaluation_name": "Japanese",
307
+ "source_data": {
308
+ "dataset_name": "global-mmlu-lite",
309
+ "source_type": "url",
310
+ "url": [
311
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
312
+ ]
313
+ },
314
+ "metric_config": {
315
+ "evaluation_description": "Global MMLU Lite - Japanese",
316
+ "lower_is_better": false,
317
+ "score_type": "continuous",
318
+ "min_score": 0.0,
319
+ "max_score": 1.0
320
+ },
321
+ "score_details": {
322
+ "score": 0.87,
323
+ "uncertainty": {
324
+ "confidence_interval": {
325
+ "lower": -0.033,
326
+ "upper": 0.033,
327
+ "method": "unknown"
328
+ }
329
+ }
330
+ }
331
+ },
332
+ {
333
+ "evaluation_name": "Korean",
334
+ "source_data": {
335
+ "dataset_name": "global-mmlu-lite",
336
+ "source_type": "url",
337
+ "url": [
338
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
339
+ ]
340
+ },
341
+ "metric_config": {
342
+ "evaluation_description": "Global MMLU Lite - Korean",
343
+ "lower_is_better": false,
344
+ "score_type": "continuous",
345
+ "min_score": 0.0,
346
+ "max_score": 1.0
347
+ },
348
+ "score_details": {
349
+ "score": 0.895,
350
+ "uncertainty": {
351
+ "confidence_interval": {
352
+ "lower": -0.03,
353
+ "upper": 0.03,
354
+ "method": "unknown"
355
+ }
356
+ }
357
+ }
358
+ },
359
+ {
360
+ "evaluation_name": "Portuguese",
361
+ "source_data": {
362
+ "dataset_name": "global-mmlu-lite",
363
+ "source_type": "url",
364
+ "url": [
365
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
366
+ ]
367
+ },
368
+ "metric_config": {
369
+ "evaluation_description": "Global MMLU Lite - Portuguese",
370
+ "lower_is_better": false,
371
+ "score_type": "continuous",
372
+ "min_score": 0.0,
373
+ "max_score": 1.0
374
+ },
375
+ "score_details": {
376
+ "score": 0.8725,
377
+ "uncertainty": {
378
+ "confidence_interval": {
379
+ "lower": -0.0327,
380
+ "upper": 0.0327,
381
+ "method": "unknown"
382
+ }
383
+ }
384
+ }
385
+ },
386
+ {
387
+ "evaluation_name": "Spanish",
388
+ "source_data": {
389
+ "dataset_name": "global-mmlu-lite",
390
+ "source_type": "url",
391
+ "url": [
392
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
393
+ ]
394
+ },
395
+ "metric_config": {
396
+ "evaluation_description": "Global MMLU Lite - Spanish",
397
+ "lower_is_better": false,
398
+ "score_type": "continuous",
399
+ "min_score": 0.0,
400
+ "max_score": 1.0
401
+ },
402
+ "score_details": {
403
+ "score": 0.9075,
404
+ "uncertainty": {
405
+ "confidence_interval": {
406
+ "lower": -0.0284,
407
+ "upper": 0.0284,
408
+ "method": "unknown"
409
+ }
410
+ }
411
+ }
412
+ },
413
+ {
414
+ "evaluation_name": "Swahili",
415
+ "source_data": {
416
+ "dataset_name": "global-mmlu-lite",
417
+ "source_type": "url",
418
+ "url": [
419
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
420
+ ]
421
+ },
422
+ "metric_config": {
423
+ "evaluation_description": "Global MMLU Lite - Swahili",
424
+ "lower_is_better": false,
425
+ "score_type": "continuous",
426
+ "min_score": 0.0,
427
+ "max_score": 1.0
428
+ },
429
+ "score_details": {
430
+ "score": 0.91,
431
+ "uncertainty": {
432
+ "confidence_interval": {
433
+ "lower": -0.028,
434
+ "upper": 0.028,
435
+ "method": "unknown"
436
+ }
437
+ }
438
+ }
439
+ },
440
+ {
441
+ "evaluation_name": "Yoruba",
442
+ "source_data": {
443
+ "dataset_name": "global-mmlu-lite",
444
+ "source_type": "url",
445
+ "url": [
446
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
447
+ ]
448
+ },
449
+ "metric_config": {
450
+ "evaluation_description": "Global MMLU Lite - Yoruba",
451
+ "lower_is_better": false,
452
+ "score_type": "continuous",
453
+ "min_score": 0.0,
454
+ "max_score": 1.0
455
+ },
456
+ "score_details": {
457
+ "score": 0.905,
458
+ "uncertainty": {
459
+ "confidence_interval": {
460
+ "lower": -0.0287,
461
+ "upper": 0.0287,
462
+ "method": "unknown"
463
+ }
464
+ }
465
+ }
466
+ },
467
+ {
468
+ "evaluation_name": "Chinese",
469
+ "source_data": {
470
+ "dataset_name": "global-mmlu-lite",
471
+ "source_type": "url",
472
+ "url": [
473
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
474
+ ]
475
+ },
476
+ "metric_config": {
477
+ "evaluation_description": "Global MMLU Lite - Chinese",
478
+ "lower_is_better": false,
479
+ "score_type": "continuous",
480
+ "min_score": 0.0,
481
+ "max_score": 1.0
482
+ },
483
+ "score_details": {
484
+ "score": 0.8525,
485
+ "uncertainty": {
486
+ "confidence_interval": {
487
+ "lower": -0.0348,
488
+ "upper": 0.0348,
489
+ "method": "unknown"
490
+ }
491
+ }
492
+ }
493
+ },
494
+ {
495
+ "evaluation_name": "Burmese",
496
+ "source_data": {
497
+ "dataset_name": "global-mmlu-lite",
498
+ "source_type": "url",
499
+ "url": [
500
+ "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite"
501
+ ]
502
+ },
503
+ "metric_config": {
504
+ "evaluation_description": "Global MMLU Lite - Burmese",
505
+ "lower_is_better": false,
506
+ "score_type": "continuous",
507
+ "min_score": 0.0,
508
+ "max_score": 1.0
509
+ },
510
+ "score_details": {
511
+ "score": 0.9075,
512
+ "uncertainty": {
513
+ "confidence_interval": {
514
+ "lower": -0.0284,
515
+ "upper": 0.0284,
516
+ "method": "unknown"
517
+ }
518
+ }
519
+ }
520
+ }
521
+ ]
522
+ }
data/rewardbench/0-hero/Matter-0.1-7B-DPO-preview/2fbdc84a-0777-4e21-9143-cb2d9c75cc20.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/0-hero_Matter-0.1-7B-DPO-preview/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "0-hero/Matter-0.1-7B-DPO-preview",
18
+ "id": "0-hero/Matter-0.1-7B-DPO-preview",
19
+ "developer": "0-hero",
20
+ "additional_details": {
21
+ "model_type": "DPO"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.7247
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.8939
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.5768
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.6378
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.8854
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.5348
131
+ }
132
+ }
133
+ ]
134
+ }
data/rewardbench/0-hero/Matter-0.1-7B-boost-DPO-preview/ca0344bb-1c2b-4796-8b85-d326fc89a2aa.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/0-hero_Matter-0.1-7B-boost-DPO-preview/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "0-hero/Matter-0.1-7B-boost-DPO-preview",
18
+ "id": "0-hero/Matter-0.1-7B-boost-DPO-preview",
19
+ "developer": "0-hero",
20
+ "additional_details": {
21
+ "model_type": "DPO"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.7448
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.9106
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.6096
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.7135
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.8395
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.5566
131
+ }
132
+ }
133
+ ]
134
+ }
data/rewardbench/Ahjeong/MMPO_Gemma_7b/66190e14-2862-4f53-8224-f968e4d55732.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/Ahjeong_MMPO_Gemma_7b/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "Ahjeong/MMPO_Gemma_7b",
18
+ "id": "Ahjeong/MMPO_Gemma_7b",
19
+ "developer": "Ahjeong",
20
+ "additional_details": {
21
+ "model_type": "DPO"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.7587
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.9693
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.614
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.7135
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.7756
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.6831
131
+ }
132
+ }
133
+ ]
134
+ }
data/rewardbench/Ahjeong/MMPO_Gemma_7b_gamma1.1_epoch3/e96f9bba-8efe-457e-a6b1-4f90b8739f17.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/Ahjeong_MMPO_Gemma_7b_gamma1.1_epoch3/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "Ahjeong/MMPO_Gemma_7b_gamma1.1_epoch3",
18
+ "id": "Ahjeong/MMPO_Gemma_7b_gamma1.1_epoch3",
19
+ "developer": "Ahjeong",
20
+ "additional_details": {
21
+ "model_type": "DPO"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.7652
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.9721
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.6338
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.7635
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.7284
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.6913
131
+ }
132
+ }
133
+ ]
134
+ }
data/rewardbench/Anthropic/claude-3-5-sonnet-20240620/25fcb524-3e9b-4d97-b84d-01986b127c6c.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/Anthropic_claude-3-5-sonnet-20240620/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "Anthropic/claude-3-5-sonnet-20240620",
18
+ "id": "Anthropic/claude-3-5-sonnet-20240620",
19
+ "developer": "Anthropic",
20
+ "additional_details": {
21
+ "model_type": "Generative"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.8417
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.9637
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.7401
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.8162
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.8469
113
+ }
114
+ }
115
+ ]
116
+ }
data/rewardbench/Anthropic/claude-3-haiku-20240307/9421228b-a52b-4b36-a28b-2ad49f2dd5cb.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/Anthropic_claude-3-haiku-20240307/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "Anthropic/claude-3-haiku-20240307",
18
+ "id": "Anthropic/claude-3-haiku-20240307",
19
+ "developer": "Anthropic",
20
+ "additional_details": {
21
+ "model_type": "Generative"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.7289
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.9274
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.5197
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.7953
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.706
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.6635
131
+ }
132
+ }
133
+ ]
134
+ }
data/rewardbench/Anthropic/claude-3-opus-20240229/6595e39a-1c58-424f-acc5-c6bff657bc39.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/Anthropic_claude-3-opus-20240229/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "Anthropic/claude-3-opus-20240229",
18
+ "id": "Anthropic/claude-3-opus-20240229",
19
+ "developer": "Anthropic",
20
+ "additional_details": {
21
+ "model_type": "Generative"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.8008
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.9469
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.6031
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.8662
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.7868
113
+ }
114
+ }
115
+ ]
116
+ }
data/rewardbench/Anthropic/claude-3-sonnet-20240229/473f6ca5-f833-4cec-8395-88172151dfc5.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/Anthropic_claude-3-sonnet-20240229/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "Anthropic/claude-3-sonnet-20240229",
18
+ "id": "Anthropic/claude-3-sonnet-20240229",
19
+ "developer": "Anthropic",
20
+ "additional_details": {
21
+ "model_type": "Generative"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.7458
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.9344
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.5658
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.8169
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.6907
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.6963
131
+ }
132
+ }
133
+ ]
134
+ }
data/rewardbench/AtlaAI/Selene-1-Mini-Llama-3.1-8B/d1f55fa2-64ba-481d-9bff-f82a6dd0ca45.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/AtlaAI_Selene-1-Mini-Llama-3.1-8B/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "AtlaAI/Selene-1-Mini-Llama-3.1-8B",
18
+ "id": "AtlaAI/Selene-1-Mini-Llama-3.1-8B",
19
+ "developer": "AtlaAI",
20
+ "additional_details": {
21
+ "model_type": "Generative"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.8913
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.9358
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.7939
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.8926
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.9429
113
+ }
114
+ }
115
+ ]
116
+ }
data/rewardbench/AtlaAI/Selene-1/fd470f89-4dc5-4a7c-aa6b-135d05a9aab6.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/AtlaAI_Selene-1/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "AtlaAI/Selene-1",
18
+ "id": "AtlaAI/Selene-1",
19
+ "developer": "AtlaAI",
20
+ "additional_details": {
21
+ "model_type": "Generative"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.9241
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.9777
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.8399
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.9216
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.9572
113
+ }
114
+ }
115
+ ]
116
+ }
data/rewardbench/CIR-AMS/BTRM_Qwen2_7b_0613/216e0174-bab4-40ee-9871-1be675d27445.json ADDED
@@ -0,0 +1,152 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench-2/CIR-AMS_BTRM_Qwen2_7b_0613/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench 2",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "CIR-AMS/BTRM_Qwen2_7b_0613",
18
+ "id": "CIR-AMS/BTRM_Qwen2_7b_0613",
19
+ "developer": "CIR-AMS",
20
+ "additional_details": {
21
+ "model_type": "Seq. Classifier"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench 2",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench-2-results"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.5736
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Factuality",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench 2",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench-2-results"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Factuality score - measures factual accuracy",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.5347
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Precise IF",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench 2",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench-2-results"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Precise Instruction Following score",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.3563
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Math",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench 2",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench-2-results"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Math score - measures mathematical reasoning",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.6066
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Safety",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench 2",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench-2-results"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Safety score - measures safety awareness",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.7178
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Focus",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench 2",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench-2-results"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Focus score - measures response focus",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.5737
131
+ }
132
+ },
133
+ {
134
+ "evaluation_name": "Ties",
135
+ "source_data": {
136
+ "dataset_name": "RewardBench 2",
137
+ "source_type": "hf_dataset",
138
+ "hf_repo": "allenai/reward-bench-2-results"
139
+ },
140
+ "metric_config": {
141
+ "evaluation_description": "Ties score - ability to identify tie cases",
142
+ "lower_is_better": false,
143
+ "score_type": "continuous",
144
+ "min_score": 0.0,
145
+ "max_score": 1.0
146
+ },
147
+ "score_details": {
148
+ "score": 0.6527
149
+ }
150
+ }
151
+ ]
152
+ }
data/rewardbench/CIR-AMS/BTRM_Qwen2_7b_0613/3c1e3213-e5ac-47b5-b54a-8d1cbab56e01.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/CIR-AMS_BTRM_Qwen2_7b_0613/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "CIR-AMS/BTRM_Qwen2_7b_0613",
18
+ "id": "CIR-AMS/BTRM_Qwen2_7b_0613",
19
+ "developer": "CIR-AMS",
20
+ "additional_details": {
21
+ "model_type": "Seq. Classifier"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.8172
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.9749
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.5724
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.9014
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.8775
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.7029
131
+ }
132
+ }
133
+ ]
134
+ }
data/rewardbench/CohereForAI/c4ai-command-r-plus/1468c630-4d3a-42b4-b215-c6a457e9d452.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/CohereForAI_c4ai-command-r-plus/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "CohereForAI/c4ai-command-r-plus",
18
+ "id": "CohereForAI/c4ai-command-r-plus",
19
+ "developer": "CohereForAI",
20
+ "additional_details": {
21
+ "model_type": "Generative"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.7057
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.9511
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.5757
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.5986
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.704
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.6924
131
+ }
132
+ }
133
+ ]
134
+ }
data/rewardbench/ContextualAI/LMUnit-llama3.1-70b/712e97f2-b937-44fd-a27e-6f080a3091da.json ADDED
@@ -0,0 +1,152 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench-2/ContextualAI_LMUnit-llama3.1-70b/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench 2",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "ContextualAI/LMUnit-llama3.1-70b",
18
+ "id": "ContextualAI/LMUnit-llama3.1-70b",
19
+ "developer": "ContextualAI",
20
+ "additional_details": {
21
+ "model_type": "Generative RM"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench 2",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench-2-results"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.8054
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Factuality",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench 2",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench-2-results"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Factuality score - measures factual accuracy",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.8463
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Precise IF",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench 2",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench-2-results"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Precise Instruction Following score",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.4875
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Math",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench 2",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench-2-results"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Math score - measures mathematical reasoning",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.7158
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Safety",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench 2",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench-2-results"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Safety score - measures safety awareness",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.9067
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Focus",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench 2",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench-2-results"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Focus score - measures response focus",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.9697
131
+ }
132
+ },
133
+ {
134
+ "evaluation_name": "Ties",
135
+ "source_data": {
136
+ "dataset_name": "RewardBench 2",
137
+ "source_type": "hf_dataset",
138
+ "hf_repo": "allenai/reward-bench-2-results"
139
+ },
140
+ "metric_config": {
141
+ "evaluation_description": "Ties score - ability to identify tie cases",
142
+ "lower_is_better": false,
143
+ "score_type": "continuous",
144
+ "min_score": 0.0,
145
+ "max_score": 1.0
146
+ },
147
+ "score_details": {
148
+ "score": 0.9063
149
+ }
150
+ }
151
+ ]
152
+ }
data/rewardbench/ContextualAI/LMUnit-qwen2.5-72b/9c657c91-2209-44ca-8bc0-48ec8d7e9bba.json ADDED
@@ -0,0 +1,152 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench-2/ContextualAI_LMUnit-qwen2.5-72b/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench 2",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "ContextualAI/LMUnit-qwen2.5-72b",
18
+ "id": "ContextualAI/LMUnit-qwen2.5-72b",
19
+ "developer": "ContextualAI",
20
+ "additional_details": {
21
+ "model_type": "Generative RM"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench 2",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench-2-results"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.8208
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Factuality",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench 2",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench-2-results"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Factuality score - measures factual accuracy",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.8716
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Precise IF",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench 2",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench-2-results"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Precise Instruction Following score",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.5437
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Math",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench 2",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench-2-results"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Math score - measures mathematical reasoning",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.7268
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Safety",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench 2",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench-2-results"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Safety score - measures safety awareness",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.9133
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Focus",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench 2",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench-2-results"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Focus score - measures response focus",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.9677
131
+ }
132
+ },
133
+ {
134
+ "evaluation_name": "Ties",
135
+ "source_data": {
136
+ "dataset_name": "RewardBench 2",
137
+ "source_type": "hf_dataset",
138
+ "hf_repo": "allenai/reward-bench-2-results"
139
+ },
140
+ "metric_config": {
141
+ "evaluation_description": "Ties score - ability to identify tie cases",
142
+ "lower_is_better": false,
143
+ "score_type": "continuous",
144
+ "min_score": 0.0,
145
+ "max_score": 1.0
146
+ },
147
+ "score_details": {
148
+ "score": 0.9014
149
+ }
150
+ }
151
+ ]
152
+ }
data/rewardbench/ContextualAI/archangel_sft-dpo_llama13b/19e04aeb-d489-48b2-a5ae-2c837f23f686.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/ContextualAI_archangel_sft-dpo_llama13b/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "ContextualAI/archangel_sft-dpo_llama13b",
18
+ "id": "ContextualAI/archangel_sft-dpo_llama13b",
19
+ "developer": "ContextualAI",
20
+ "additional_details": {
21
+ "model_type": "DPO"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.54
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.7123
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.4298
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.5649
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.4401
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.5656
131
+ }
132
+ }
133
+ ]
134
+ }
data/rewardbench/ContextualAI/archangel_sft-dpo_llama30b/4e7dbb9b-bd79-4e48-8b9a-4a19312e526a.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/ContextualAI_archangel_sft-dpo_llama30b/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "ContextualAI/archangel_sft-dpo_llama30b",
18
+ "id": "ContextualAI/archangel_sft-dpo_llama30b",
19
+ "developer": "ContextualAI",
20
+ "additional_details": {
21
+ "model_type": "DPO"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.5618
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.6927
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.4474
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.6284
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.4745
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.5705
131
+ }
132
+ }
133
+ ]
134
+ }
data/rewardbench/ContextualAI/archangel_sft-dpo_llama7b/2ffd133b-b4c2-4005-8166-42d1012b8397.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/ContextualAI_archangel_sft-dpo_llama7b/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "ContextualAI/archangel_sft-dpo_llama7b",
18
+ "id": "ContextualAI/archangel_sft-dpo_llama7b",
19
+ "developer": "ContextualAI",
20
+ "additional_details": {
21
+ "model_type": "DPO"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.5304
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.5782
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.4452
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.5203
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.5658
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.5544
131
+ }
132
+ }
133
+ ]
134
+ }
data/rewardbench/ContextualAI/archangel_sft-dpo_pythia1-4b/4a4b751a-ec7a-4d99-a0c2-f008cfc389a1.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/ContextualAI_archangel_sft-dpo_pythia1-4b/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "ContextualAI/archangel_sft-dpo_pythia1-4b",
18
+ "id": "ContextualAI/archangel_sft-dpo_pythia1-4b",
19
+ "developer": "ContextualAI",
20
+ "additional_details": {
21
+ "model_type": "DPO"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.5233
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.6397
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.3728
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.5041
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.5672
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.5427
131
+ }
132
+ }
133
+ ]
134
+ }
data/rewardbench/ContextualAI/archangel_sft-dpo_pythia12-0b/48a9c2c0-24f9-467d-9e2d-aabc0a8abdd1.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/ContextualAI_archangel_sft-dpo_pythia12-0b/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "ContextualAI/archangel_sft-dpo_pythia12-0b",
18
+ "id": "ContextualAI/archangel_sft-dpo_pythia12-0b",
19
+ "developer": "ContextualAI",
20
+ "additional_details": {
21
+ "model_type": "DPO"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.5009
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.6676
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.364
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.5432
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.4139
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.5303
131
+ }
132
+ }
133
+ ]
134
+ }
data/rewardbench/ContextualAI/archangel_sft-dpo_pythia2-8b/7594c6a6-43af-49e5-a2b0-f41448c2e273.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": "0.2.1",
3
+ "evaluation_id": "reward-bench/ContextualAI_archangel_sft-dpo_pythia2-8b/1783820827.6488361",
4
+ "retrieved_timestamp": "1783820827.6488361",
5
+ "source_metadata": {
6
+ "source_name": "RewardBench",
7
+ "source_type": "documentation",
8
+ "source_organization_name": "Allen Institute for AI",
9
+ "source_organization_url": "https://allenai.org",
10
+ "evaluator_relationship": "third_party"
11
+ },
12
+ "eval_library": {
13
+ "name": "unknown",
14
+ "version": "unknown"
15
+ },
16
+ "model_info": {
17
+ "name": "ContextualAI/archangel_sft-dpo_pythia2-8b",
18
+ "id": "ContextualAI/archangel_sft-dpo_pythia2-8b",
19
+ "developer": "ContextualAI",
20
+ "additional_details": {
21
+ "model_type": "DPO"
22
+ }
23
+ },
24
+ "evaluation_results": [
25
+ {
26
+ "evaluation_name": "Score",
27
+ "source_data": {
28
+ "dataset_name": "RewardBench",
29
+ "source_type": "hf_dataset",
30
+ "hf_repo": "allenai/reward-bench"
31
+ },
32
+ "metric_config": {
33
+ "evaluation_description": "Overall RewardBench Score",
34
+ "lower_is_better": false,
35
+ "score_type": "continuous",
36
+ "min_score": 0.0,
37
+ "max_score": 1.0
38
+ },
39
+ "score_details": {
40
+ "score": 0.5286
41
+ }
42
+ },
43
+ {
44
+ "evaluation_name": "Chat",
45
+ "source_data": {
46
+ "dataset_name": "RewardBench",
47
+ "source_type": "hf_dataset",
48
+ "hf_repo": "allenai/reward-bench"
49
+ },
50
+ "metric_config": {
51
+ "evaluation_description": "Chat accuracy - includes easy chat subsets",
52
+ "lower_is_better": false,
53
+ "score_type": "continuous",
54
+ "min_score": 0.0,
55
+ "max_score": 1.0
56
+ },
57
+ "score_details": {
58
+ "score": 0.8073
59
+ }
60
+ },
61
+ {
62
+ "evaluation_name": "Chat Hard",
63
+ "source_data": {
64
+ "dataset_name": "RewardBench",
65
+ "source_type": "hf_dataset",
66
+ "hf_repo": "allenai/reward-bench"
67
+ },
68
+ "metric_config": {
69
+ "evaluation_description": "Chat Hard accuracy - includes hard chat subsets",
70
+ "lower_is_better": false,
71
+ "score_type": "continuous",
72
+ "min_score": 0.0,
73
+ "max_score": 1.0
74
+ },
75
+ "score_details": {
76
+ "score": 0.3355
77
+ }
78
+ },
79
+ {
80
+ "evaluation_name": "Safety",
81
+ "source_data": {
82
+ "dataset_name": "RewardBench",
83
+ "source_type": "hf_dataset",
84
+ "hf_repo": "allenai/reward-bench"
85
+ },
86
+ "metric_config": {
87
+ "evaluation_description": "Safety accuracy - includes safety subsets",
88
+ "lower_is_better": false,
89
+ "score_type": "continuous",
90
+ "min_score": 0.0,
91
+ "max_score": 1.0
92
+ },
93
+ "score_details": {
94
+ "score": 0.4473
95
+ }
96
+ },
97
+ {
98
+ "evaluation_name": "Reasoning",
99
+ "source_data": {
100
+ "dataset_name": "RewardBench",
101
+ "source_type": "hf_dataset",
102
+ "hf_repo": "allenai/reward-bench"
103
+ },
104
+ "metric_config": {
105
+ "evaluation_description": "Reasoning accuracy - includes code and math subsets",
106
+ "lower_is_better": false,
107
+ "score_type": "continuous",
108
+ "min_score": 0.0,
109
+ "max_score": 1.0
110
+ },
111
+ "score_details": {
112
+ "score": 0.5135
113
+ }
114
+ },
115
+ {
116
+ "evaluation_name": "Prior Sets (0.5 weight)",
117
+ "source_data": {
118
+ "dataset_name": "RewardBench",
119
+ "source_type": "hf_dataset",
120
+ "hf_repo": "allenai/reward-bench"
121
+ },
122
+ "metric_config": {
123
+ "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets",
124
+ "lower_is_better": false,
125
+ "score_type": "continuous",
126
+ "min_score": 0.0,
127
+ "max_score": 1.0
128
+ },
129
+ "score_details": {
130
+ "score": 0.5501
131
+ }
132
+ }
133
+ ]
134
+ }